Good day, and welcome to the Aehr Test Systems fiscal 2020 fourth quarter and full year financial results call. Today's conference is being recorded. At this time, I would like to turn the conference over to Mr. Jim Byers of the MKR Group. Please go ahead, sir.
Thank you, operator. Good afternoon, and welcome to Aehr Test Systems' fiscal 2020 fourth quarter and full year financial results conference call. With me on today's call are Aehr Test Systems' President and Chief Executive Officer, Gayn Erickson, and Chief Financial Officer, Ken Spink. Before I turn the call over to Gayn and Ken, I'd like to cover a few quick items this afternoon. Aehr Test issued a press release announcing its fiscal 2020 fourth quarter and full year results. That release is available on the company's website at aehr.com. This call is being broadcast live over the internet for all interested parties, and the webcast will be archived on the investor relations page of the company's website.
I'd like to remind everyone that on today's call, management will be making forward-looking statements today that are based on current information and estimates and are subject to a number of risks and uncertainties that could cause actual results to differ materially from those in the forward-looking statements. These factors that may cause results to differ materially from those in the forward-looking statements are discussed in the company's most recent periodic and current reports filed with the SEC. These forward-looking statements, including guidance provided during today's call, are only valid as of this date, and Aehr Test Systems undertakes no obligation to update the forward-looking statements. Now, with that said, I'd like to turn the call over to Gayn Erickson, President and Chief Executive Officer. Gayn?
Thanks, Jim, good afternoon to those joining us on today's conference call and also listening online. Ken will go over our fourth quarter and full year financial results later in the call, but first, I'll spend a few minutes discussing our business and product highlights, including our continued progress with our wafer level and simulated die testing burn-in solutions. Then we'll open it up for your questions. This past fiscal year, we made substantial progress with our new FOX products that strengthened our customer base, expanded our markets, and enhanced our operations and sales capabilities to capitalize on the significant market opportunities we see ahead.
While we were on track to meet our expected guidance for fiscal 2020, due to the challenging global environment and uncertainty around the COVID-19 pandemic, we experienced pushouts of customer forecasted orders in our second fiscal half for our FOX-S systems and consumables in data center and some 5G end use applications for silicon photonics transceivers. These customers have indicated the pushouts are temporary and that they'll require the additional system capacity and consumables in the current fiscal 2021 year. We'll cover the details supporting our optimism but want to quickly state that we are reinstating guidance and expect FY 2021 full year revenue to be between $25 and $28 million, up 12%-26%, and to be profitable for the year.
With the increase in number of customers in production using our systems, the new market opportunities we added with new customers this year, our move to our higher margin FOX systems and consumables, and the completion of our previously announced restructuring and sales enhancements this past year, going forward, we are well-positioned to address our new market opportunities and are now profitable at a much lower revenue level. Let me walk you through some of the key business highlights for this last quarter and for the last fiscal year, as we outlined in our earnings release. First highlight, we closed a new order with a major new customer in silicon photonics.
During the quarter, Aehr closed an initial order with a new customer that is a major global leader of communication transceivers for data centers, telecom, and 5G infrastructure for our FOX full wafer level test and burn-in system for production stabilization and test of their Silicon Photonics devices. This new customer is deploying our FOX-NP system for initial production burn-in and stabilization of their high-performance Silicon Photonics devices and is forecasted to then transition to our FOX-XP wafer level test and burn-in systems during this fiscal year 2021 to meet their volume production forecast. We categorize this customer as a tier 1 customer, which we define as a customer with the resources and market size to be able to purchase $6 million-$10 million per year or more of our systems and consumables. The next highlight is we closed an initial order with the world's largest OSAT.
During the quarter, we closed an initial order with the world's largest outsourced semiconductor assembly and test supplier to use the FOX-P family of products, including Aehr wafer packs and die packs, for production test, burn-in, and reliability screening of devices at full wafer simulated die and module. They have already added our system to their list of tools and capabilities in their marketing and sales material to their customers. We have begun some cross-marketing and sales activities with them. Stay tuned for more updates and announcements. Next, Aehr added a key new market with the addition of wafer level burn-in of Silicon Carbide devices.
This year, we successfully took the initial order for and installed our first production capacity for Silicon Carbide devices, including it to the list of markets such as silicon photonics, 2D and 3D sensors, automotive lasers used in photonics devices, and have shown the value and feasibility of using our FOX solutions to address these market needs. The initial system order was for a FOX multi-wafer system with proprietary wafer packs configured to test 18 wafers in parallel at up to 1,000 watts of power per wafer.
The customer is using it for 100% production burn-in and infant mortality screening of Silicon Carbide devices at wafer level. This new Silicon Carbide application with a Fortune 500 market leader in Silicon Carbide and power modules adds a significant new Tier 1 customer of our FOX-XP system and wafer packs for whole wafer burn-in and infant mortality screening of Silicon Carbide devices. Since our initial installation in January, we've received multiple follow-on orders for additional wafer packs from this customer, including multiple new designs, and now have a significant number of different devices that have been released into production. This customer is forecasting additional capacity needs for our FOX-XP systems during this fiscal year and for years into the future.
The Silicon Carbide market, semiconductor device market, is growing at a tremendous rate, with a unit growth of high-power devices of over 50% CAGR per Yole Développement research from 2019 to 2025. Silicon Carbide is a very impressive material for high power, and particularly high voltage devices for applications such as the needs of electric and hybrid electric vehicle powertrains, electric vehicle charging infrastructure, IT data center power supplies, and renewable energy power conversion such as wind, solar, as well as power storage. These devices have shown reduction in power losses as much as 78%. Many articles have been written about the first mainstream use of Silicon Carbide power devices that were in the Tesla Model 3, which enabled much longer driving range per charge.
This has basically changed the market, with most, if not every electric vehicle, hybrid electric vehicle, automotive company, moving to Silicon Carbide-based powertrain and charging system. The challenge is the reliability. Silicon Carbide is known to have high infant mortality rates, but after a reliability burn-in screening, these defects can be completely removed to provide extremely reliable devices for these mission-critical applications. Aehr is able to provide a complete solution for one of the key reliability screening tests on an entire wafer of devices all at one time, while testing and monitoring every device for failures during the burn-in process to provide critical information on devices so they're not later packaged into multi-die modules where the yield impact is 10x or 100 times as costly.
Yole Développement research has forecast of over 600 million yielded power MOSFET 20 amp equivalent devices shipping per year by 2025, equates to over a half a million wafer starts per year, which creates an enormous opportunity for our wafer level and singulated die systems, given the long durations required to burn in the devices and to remove the defective parts. Burn-in times can be as long as days per wafer, even at our industry-leading 18 wafer per system capacity, FOX-XP, that's a significant number of systems. The next highlight is pretty important. In fiscal 2020, we saw the industry adopt production wafer level burn-in.
We made significant progress with our new FOX products for wafer and singulated die test and burn-in during the fiscal year, with two Tier 1 customers added and five customers transitioning to production with 100% stabilization or infant mortality screening with our FOX systems. We saw our silicon photonics customers move to production for the first time in just this last fiscal 2020. During the fiscal year, we saw our lead customers for silicon photonics move to full volume production. We expect them to purchase additional systems this fiscal year and into the future as they continue to maintain or grow their market share and add additional silicon photonics devices to the mix. We also moved three additional silicon photonics customers to production with our FOX systems in fiscal 2020.
All three of these customers are expected to ramp production during this fiscal 2021 as well, adding capacity in both systems and consumables. Near the fiscal year-end, we announced yet another new silicon photonics customer that is deploying our FOX-NP system for initial production burn-in and stabilization of their high-performance silicon photonics devices, and is forecasted to then transition to our FOX-XP multi-wafer systems during the fiscal 2021 to meet their volume production forecast. Silicon photonics devices are highly integrated silicon-based semiconductors that have embedded or integrated the non-silicon-based laser transmitters and receivers to enable a smaller, lower cost, higher reliable alternative to traditional fiber optic transceivers. Historically, fiber optic transceivers are made up of many different logic ICs, multiplexers, demuxes, external discrete lasers, and receivers into a mechanical package that is used in data center and telecommunication infrastructure.
This has been the high-speed transmission lines for long haul and data center to data center backplane of the internet. These fiber optic transceivers have been extremely difficult and expensive to build. This has been seen as a limiter to the adoption of fiber optic transmission of data and to the maximum data rates and transmission in the data centers that store the world's data. Yole Développement research has stated that market leaders like Intel, Cisco, Luxtera, Padock, Inphi, and Acacia are setting the standards for 100, 200, 400, and even 800 gigabit transceiver standards based on transceivers with fully integrated silicon photonics devices, while many other companies are also jumping into this exploding market.
One of the key claims of these transceivers are their lower manufacturing cost and the ability to scale manufacturing due to the full wafer level integration of these devices, which brings the scale of semiconductor manufacturing to fiber optic communication for the first time in history. Where Aehr fits in is that these devices all need to have their photonics transmitter stabilized under high power and temperature. Also customers use our systems to screen for infant mortality of these devices to ensure high initial quality and long-term reliability. This is a manufacturing step done on 100% of the die. In the case of silicon photonics, we provide a much more cost-effective and scalable solution for this step than doing this equivalent stabilization and screening after the die are put into the final PCB substrate and package.
The silicon photonics market is growing at a CAGR of 42% between 2019 and 2025 to a $3.6 billion annual market. We believe that the entire industry will transition to wafer level or simulated die for this critical manufacturing step, which is where our FOX-P products stand alone as the most cost-effective solution for this and a portfolio of patents and IP in this area. We estimate that the market opportunity for wafer stabilization, reliability screening equipment, and contactors for silicon photonics is approximately $150 million by 2025, with well over 300 wafers of test capacity required by that time. Our FOX-XP production system is the only multi-wafer system available to test and burn in these high-power silicon photonics wafers in a single insertion, and we can test up to nine 2,000-watt wafers in parallel on a single system.
The total capacity needed by 2025 is about 35 of our nine-wafer FOX-XP systems, to put this into perspective. To date, Aehr has shipped about 50 wafers of capacity into this application. Interestingly, while fiscal 2020 was the first year to see volume production of silicon photonics with wafer level burn-in, FY 2020 also saw second half pushouts in silicon photonics ramps. We experienced pushouts in customer forecasted orders in our second fiscal half of fiscal 2020 for our FOX systems and consumables in data center and some 5G end-use applications for silicon photonics transceivers. These customers, as I said before, have indicated the pushouts are temporary and they require the additional system capacity and consumables in the current fiscal 2021 year. Our next highlight is shipments of consumables were a significant percentage of revenue this year.
Shipments of our proprietary wafer pack contactors and die pack carrier consumables for our FOX systems accounted for 48% of total revenue in fiscal 2020. In fiscal Q4, which we just ended, our consumable revenue was 79% of revenue. As anticipated, customer orders for systems did not materialize, but customer demand for the consumables for the installed base systems held steady. As we stated in the past, historically, consumables can often soften any weakness in systems as customers contemplate new capacity but maintain, in some cases, actually increase the need for new wafer pack contactors and die pack carriers to get new designs or devices out to market.
In the semiconductor test industry, which was just over $9 billion total last year in 2019, it is made up of $3.7 billion in test systems, another $3.7 billion in consumables, such as probe cards for contacting wafers and sockets and pack boards for contacting package parts, another $1.7 billion is semiconductor device handling equipment in wafer and package form. With Aehr's FOX product line, we actually play in all three segments. Our FOX systems serve the test systems market. Our wafer packs and die packs serve the contactor consumables market, our aligners and the FOX systems themselves that have the integrated thermal capabilities of a wafer prober are a turnkey solution for handling devices. The consumable business as a whole is approximately the same size and often higher than the systems business in down years in the overall semiconductor test business.
Both about $3.7 billion. For reliability and burn-in space, which we primarily play in, the consumables can be two to four times the annual sale of systems, as the systems typically are used for longer periods of time with annual needs for new contactors and consumables. This is why we're confident that our consumable business is likely to exceed our overall systems business over time, even though both will grow in absolute dollars. Our next highlight is that Aehr is currently engaged with over a dozen new potential customers. We're currently working with well over a dozen additional tier 1 and tier 2 customers that are considering using our products for high market growth applications, including silicon photonics,S ilicon Carbide, automotive, and memory devices production burn-in.
While tier 1 customers are seen as those with the opportunity to drive $6 million-$10 million or more in systems and consumables per year, our tier 2 customers are considered to have the market share and application to drive $1 million-$3 million per year or more. Several of these companies are expected to place their orders this year with ramps into production later in the fiscal year and or the following fiscal year. We see an increasing awareness and adoption rate that we believe could drive the majority of the market for Silicon Carbide as well as silicon photonics to move to wafer level or simulated die burn-in within the next few years. Our final highlight is in FY 2020, we completed our planned restructuring and shift to higher margin products.
Aehr completed our previously announced restructuring and also moved to much higher margin FOX systems and consumables during the fiscal year. We started this before the pandemic outbreak and completed it during the last few months. As part of the previously announced and planned restructuring, we completed the closure of our Japan subsidiary and also transitioned our European sales to third-party sales representatives late in the fiscal year. We also added key marketing directors and made some additional structural changes to our sales force. We believe these enhancements have already and will continue to both improve our efficiency and materially increase our sales activity and bookings going forward and increase our penetration of key customers in our target markets. We believe these changes position us for success for sales of our current products, as well as additional new products planned for introduction this year.
We also have shifted to higher margin, highly differentiated systems and consumables. As I noted in the last call, we've started to see some forecasts for renewed market demand for packaged part burn-in systems, particularly from customers who are asking us about our high voltage capability and adding this capability to our packaged part systems. These changes in long-term forecasts reflect the move toward higher voltages and other requirements for devices in automobiles, particularly with electric and hybrid automobiles and autonomous vehicle sensors. We expect to see a resurgence of packaged part burn-in systems orders from some specific ABTS system customers, and to generate additional new opportunities with our planned introduction this fiscal year of a new packaged part burn-in system product that adds very high voltage test capability.
We see the need for high voltage capabilities in both wafer level and packaged part as a new high-growth opportunity for Aehr Test, and expect to see sales from current customers resume, and also add several new customers that include both tier 1 and tier 2 level customers for packaged part burn-in. At the same time, and as discussed last year, we had seen a significant drop-off in our packaged part product business as several of our customers have shifted their businesses or entirely closed product lines that were driving the need for test and burn-in using the high power and high pin count capabilities of our ABTS family of products.
In one specific case, a customer that had been buying multiple systems per year has all but shuttered a line of products. We feel they are unlikely to take additional capacity of that particular configuration of system that we had several systems of inventory left on-hand when they dropped their forecast to 0. Interestingly, this and several other customers have, at the same time, shifted their focus to other product lines, particularly for automotive and other new applications that are expected to drive new needs in the future. I mentioned this specifically last quarter and noted that as a result, we were going to do a deep dive in our inventory for older products and configurations.
We decided it was prudent and appropriate to write down the inventory that we simply do not see a likelihood of selling in the foreseeable future, which resulted in a one-time charge this quarter of $1.6 million of inventory. This leaves us with significant inventory of systems and material that is in our near-term forecast, particularly in our FOX products, which also allows us to make short lead time shipments, as well as meeting significant revenue forecasts without taking on additional inventory expenditures. Last year, we reported on the shipment of our new FOX-CP test and burn-in system to a major new tier 1 customer for a very high volume application for the enterprise and data center market, with a planned build-out of this production ramp over the next several years.
The FOX-CP is our low-cost single wafer compact test and reliability verification solution for logic memory and photonic devices. Their solution is comprised of a test system integrated with a wafer probe or configured with a high-powered thermal chuck that allows up to 2 kilowatts of testing or burning in of full wafers. This customer has indicated they plan to begin their production ramp within our current fiscal year. We expect to begin additional shipments of test cells to them in the second half of this fiscal year. We're very excited about this application, which is expected to drive very high volumes of devices that we believe will drive test system sales for several years. Let me try and wrap this up. We added two key tier 1 customers this past year. We now have five significantly large tier 1 customers.
Again, applications and market sizes that could drive $6 million-$10 million or more a year on our FOX wafer level singulated die test systems and consumables. We also have another seven tier 2 customers that are each capable of FOX product sales, typically between $1 million-$3 million and sometimes more. In addition, five of our customers moved to production during the year using our FOX products for 100% stabilization and burn-in and infant mortality. We will be growing the list of both tier 1 and tier 2 customers this year in both wafer level, singulated die, but also some packaged part markets, and feel we can significantly grow these and new customers in the markets we're already serving. Additionally, we will be adding new markets and enhancements to address some significantly large new markets later this fiscal year.
We're also seeing renewed activity and interest of our FOX systems and consumables for several new applications in the 2D and 3D sensor markets, particularly for mobile devices. These sensors are becoming ubiquitous in smartphones, tablets, and are forecasted to be adopted in laptops and computers as well. The level of security associated with facial recognition far exceeds fingerprint-based biometrics, and certainly greater than traditional keyboard entry passwords. These new opportunities in 2D and 3D sensing are opportunities that could add significant upside to our currently forecasted revenue for this year and next, but are not built into our current guidance.
Although COVID-19 has created challenges such as international travel, some small impacts on our supply chain, and created caution and/or delays with some customer production ramps, we believe that there is no long-term negative impact to Aehr, the demand for our products, or for the attractiveness of the key markets that we serve. We absolutely believe that we'll come out of this stronger than we went into this worldwide pandemic, with more production customers, more applications, and higher margins with higher value products.
Our key customers' products are being used to build out new data centers, improve data rates and increase storage in data centers, build out the 5G infrastructure, enable the newest sensors and technology in smartphones and tablets. Enable the widespread adoption of electric and hybrid electric vehicles and charging stations, and address the unstoppable demand for memory and data storage and computing data centers, mobile devices, and hundreds of applications that are keeping the world connected. As we move into fiscal 2021, we remain optimistic about both growth in systems and consumables within our installed base of customers, as well as expanding the number of customers with our family of FOX solutions. We expect significant growth in both our top and bottom lines moving forward, with much lower fixed operating expenses and significantly higher margin products and services.
With that, let me turn it over to Ken before we open up the line for questions.
Thank you, Gayn, and good afternoon, everyone. As Gayn mentioned, our fourth quarter results reflected the impact of the current challenging global business environment around the COVID pandemic and customers who pushed out forecasted orders during the second half of fiscal 2020. In our prior year call, we announced our two new FOX-NP customers and our new FOX-CP customers were expected to ramp to volume production during fiscal 2020 and add capacity, resulting in orders for FOX-XP systems and CP systems. While these orders did not happen in the fiscal 2020 year just reported, as their timing was pushed out by the customer, we are confident that these orders will occur in our current 2021 fiscal year.
It is important to note that through the third quarter of fiscal 2020, the company recognized revenues of $18.5 million or a little over $6 million per quarter and was profitable. This reflects the impact of the cost reduction initiatives announced in the prior year, effective fiscal 2020, and a change in product mix, which allows the company to be profitable at lower revenue levels. In the fourth quarter of fiscal 2020, revenues decreased significantly as there were no system revenues recognized during the quarter. Let me take you through our results. Net sales in the fourth quarter were $3.8 million, compared to $6.1 million in the preceding third quarter and $7.2 million in the fourth quarter of the previous year.
The decrease from Q3 includes a decrease in wafer-level burn-in revenues of $2.2 million, primarily due to a decrease in wafer-level burn-in system revenues of $2.1 million. The decrease from Q4 last year includes a decrease in wafer-level burn-in revenues of $3.3 million, primarily due to a decrease of $3.6 million in wafer-level burn-in system revenues, and a decrease in customer service revenues of $206,000. Non-GAAP net loss for the fourth quarter was $720,000 or $0.03 per diluted share, compared to non-GAAP net income of $452,000 or $0.02 per diluted share in the preceding quarter, and a non-GAAP net income of $428,000 or $0.02 per diluted share in the fourth quarter of the previous year.
The non-GAAP results exclude the impact of stock-based compensation expense, restructuring charges, and write-down of excess and obsolete inventory. On a GAAP basis, net loss for the fourth quarter was $2.9 million or $0.13 per diluted share, which includes the impact of approximately $1.9 million or $0.08 per share in inventory write-down and restructuring charges taken in the quarter. This compares to GAAP net income of $245,000 or $0.01 per diluted share in the preceding quarter, and GAAP net income of $110,000 or $0.00 per diluted share, which includes the impact of $118,000 or $0.01 per share in restructuring charges in the fourth quarter for the prior year.
Excess and obsolete inventory reserves of $1.6 million were taken in Q4 2020, related to slow-moving and obsolete packaged part burn-in product inventory, legacy FOX inventory, and down-rev FOX-P products and subsystems. The $220,000 restructuring charge consisted of severance payments and associated legal fees for individuals impacted by the closure of our subsidiary in Japan and a reduction in headcount in our German subsidiary. As noted in our last call, we will be moving to a sales rep distributorship model for sales in these regions. Gross loss in the fourth quarter was $93,000 or 2% of sales, compared to a gross profit of $3 million or 49% of sales in the preceding third quarter, and gross profit of $3.4 million or 47% of sales in the fourth quarter of the previous year.
The sequential and year-over-year decrease in gross margin is primarily due to the impact of the $1.6 million excess and obsolete inventory provision in Q4, accounting for a 44 percentage point margin impact during the quarter. In addition, gross margin was unfavorably impacted due to higher unabsorbed overhead costs to cost of sales due to lower revenue levels in the fourth quarter. As we've noted on prior calls, our wafer pack and die pack revenues are accounting for a more significant portion of our overall revenues, favorably impacting our gross margins as they maintain higher margins in our system or pass-through products. In the fourth quarter, our wafer pack and die pack consumable business accounted for 79% of total revenues, up from 51% in the preceding Q3 and 36% in Q4 of last year, providing a favorable direct material margin impact for the company.
Operating expenses in the fourth quarter were $2.7 million, flat compared to the preceding third quarter, and down $508,000 from $3.3 million in the fourth quarter last year. While Q4 2020 was flat to the preceding quarter, a decrease in SG&A of $217,000 was partially offset by the restructuring charges of $220,000 taken in the quarter. The decrease in operating expenses from prior year fourth quarter is primarily due to restructuring actions taken in the prior year related to reduced costs. SG&A was $1.7 million for the fourth quarter, compared to $1.9 million the preceding third quarter, and $2 million in the prior year fourth quarter. The decrease from Q3 is due primarily to lower labor-related costs, including lower salary expense and lower commissions from lower bookings and sales incentives in the fourth quarter.
Savings resulted to closure of our Japan subsidiary, and lower travel due to travel restrictions related to COVID-19. The decrease from prior year fourth quarter is primarily due to the impact of cost reduction initiatives in fiscal 2019. R&D expenses were $854,000 in the fourth quarter, flat compared to $845,000 the preceding third quarter, and down $266,000 from $1.1 million the prior year fourth quarter. The decrease in R&D expenses from Q4 last year is primarily due to cost reduction initiatives in fiscal 2019, and lower R&D project materials. Turning to results for the full fiscal year. Net sales for fiscal 2020 were $22.3 million, up 6% from net sales of $21.1 million in fiscal 2019. The increase includes an increase in wafer level burn-in revenues of $5.3 million, partially offset by a decrease in packaged parts revenues of $2.2 million, and customer service revenues of $1.9 million.
The decrease in packaged parts revenues is due to no ABTS system revenue in FY 2020, compared to 3 ABTS systems sold in FY 2019. Fiscal 2020 net sales were comprised of $18.9 million in wafer level burn-in revenue, and $3.4 million in customer service revenue. For the full fiscal 2020, system revenues accounted for 36% of revenues, compared to 45% in fiscal 2019. Wafer pack and die pack consumable revenues accounted for 48% of total revenue in FY 2020, compared to 29% of revenues in FY 2019. Customer service revenues accounted for 15% of revenues in FY 2020, compared to 25% of revenues in FY 2019. Non-GAAP net loss for fiscal 2020 was $27,000, or $0.00 per diluted share, compared to a non-GAAP net loss of $2.8 million, or 13% per diluted share in fiscal 2019. While revenues increased by $1.2 million FY 2020 compared to FY 2019, the non-GAAP bottom line improved by $2.8 million.
As noted earlier, this reflects the impact of cost reductions initiatives effective in fiscal 2020, and the change in product mix, which allowed the company to be profitable at lower revenue levels. On a GAAP basis, net loss for the fiscal year was $2.8 million, or $0.12 per diluted share, which includes the impact of approximately $1.9 million, or $0.08 per share, in inventory write-down and restructuring charges taken in Q4. This compares to a GAAP net loss of $5.2 million, or $0.23 per diluted share, which includes the impact of $1.5 million, or $0.07 per share, inventory write-down and restructuring charges taken in fiscal 2019. Gross profit for fiscal 2020 was $8.4 million, or 38% of net sales, compared to a gross profit of $7.6 million, or 36% of net sales in fiscal 2019.
The increase in gross margin percentage in FY 2020 compared to the prior year is primarily due to lower direct material costs due to a change in product mix, partially offset by the impact of excess and obsolete charges in FY 2020 compared to FY 2019. Operating expenses for fiscal 2020 were $11.1 million, a decrease of $1.5 million, or 12%, from $12.6 million in fiscal 2019. The decrease included a decrease in the SG&A of $194,000, a decrease in R&D of $767,000, and a decrease in restructuring charges of $505,000. SG&A was $7.5 million in fiscal 2020, down from $7.7 million reported in fiscal 2019, primarily due to cost reduction initiatives in fiscal 2019. R&D expenses were $3.4 million in fiscal 2020, down from $4.2 million in fiscal 2019. The decrease is primarily due to a cost reduction initiatives in fiscal 2019 and lower R&D material expenses.
Overall, the decrease in SG&A and R&D of $961,000 included $922,000 decrease in labor-related costs as a result of cost reduction initiatives implemented. Turning to the balance sheet for the fourth quarter. Our cash and cash equivalents were $5.4 million, up $375,000 compared to $5.1 million at the end of the preceding quarter. Accounts receivable at quarter end was $3.7 million, an increase of $206,000 compared to $3.5 million at the preceding quarter end, due to the impact of customer deposits and deferred revenue on current quarter revenues. Inventories at May 31st were $8 million compared to $9.3 million at the preceding quarter end. The $1.3 million decrease is primarily due to the $1.6 million E&O inventory provision taken in Q4 2020. Property and equipment was $663,000, compared to $783,000 the preceding quarter end.
Customer deposits and deferred revenue, short-term and long-term, were $192,000, a decrease of $227,000 compared to $419,000 at the preceding quarter end, primarily due to the decrease in backlog from the prior quarter.
Our current long-term debt of $1.7 million is related to funds we received during the fourth quarter under the Paycheck Protection Program, or PPP, which we announced in an 8-K filing in late April. We expect over $1.4 million, up to the full loan balance to be forgiven under the provisions of the CARES Act. Bookings in the fourth quarter totaled $2.6 million. Backlog at May 31st was $2.5 million compared to $3.6 million at the end of the preceding quarter, and $7.5 million at the end of the fourth quarter of the prior year. Now turning to our outlook for fiscal 2021. For our fiscal 2021 year ended May 31st, 2021, we expect full year total revenue of between $25 million and $28 million, which will represent growth between 12% and 26% year-over-year, and to be profitable for the year.
As a requirement to receive funding under the CARES Act, the company is required to use the funds from the PPP loan to retain employees and maintain payroll. While our revenues and backlog have decreased from the impact of the current challenging global business environment around the COVID pandemic, the company has made no headcount reductions. However, the company has taken actions to control spending through mandatory vacations, shutdown days, and travel restrictions. This concludes our prepared remarks. We're now ready to take your questions. Operator, please go ahead.
Thank you, sir. If you would like to ask a question, please signal by pressing star 1 on your telephone keypad. If you are using a speakerphone, please make sure your mute function is turned off to allow your signal to reach our equipment. Again, press star 1 to ask a question. We'll pause for just a moment to allow everyone an opportunity to signal for questions. As a reminder, that's star 1 to ask a question. We'll take our first question from Christian Schwab with Craig-Hallum.
Hi, This is Tyler on for Christian. Thanks for letting us ask a couple of questions.
Hey, Guys.
Hey, Guys. As we look into fiscal 2021, could you help with any expectations for linearity quarter to quarter? Within that, are you expecting maybe some softness and some customer pushoffs to continue in the first half? Any customer orders or conversations with them that would give you any indication that the first half or second half would be stronger or weaker would be helpful.
Couple things. One is, I think it's a fair, a number of the customers, and we specifically stated in the call notes, are forecasting checking out. I think we do think our second half will be stronger than our first half, and I think that's a reasonable conservative stance to take. We have not provided quarterly guidance before. Basically, we do start the quarter with a small backlog. We've got several customers with forecast for systems that we have on hand and in inventory and can turn in weeks. In some cases, they had planned to take them last quarter. We have that in the inventory that we have. Whether they come in in time to ship by the end of this quarter or next is the over-under. Honestly, we're focused on new wins and customers and markets that need our systems.
I can tell you, waking up every day and checking for expected POs doesn't help. I will also tell you I do that more often than I want to admit to. Our customers are telling their customers and their shareholders that they're ramping. In many cases, I don't even care what the customers necessarily just tell me. I'm out checking and making sure that I believe them by what are they telling the street. Most of these customers are public and large. Okay? They need our tools. We're the plan of record. That was a critical thing for us this year if we looked at our business plan, is to make sure that those wins move to production, that they actually are counting on us and get those qualified through to their end customers. Particularly things where they're automotive qualification, that's a big deal.
Basically, we think we plan to do our guidance and more without winning another customer. We have the customers, we have the applications, the products, the inventory, and we completely have the manufacturing capacity. We didn't talk much about that, but have the ability to build far in excess of even our guidance to meet customer needs, and we're adding more customers and applications, and basically that's what we're focusing on.
Thanks. That was very helpful. Second question is, you outlined you have 6 tier 1 customers who are capable of doing $6 million-$10 million a year, and you have 7 tier 2 customers who are capable of doing $1 million-$3 million a year. Obviously just math that if those customers, majority of those customers ramped into those revenue levels, your total revenue would be significantly higher than it is today. Also understanding, you have dozens more customers you're engaging with. I was just wondering how we should think about, or how you guys think about those customers ramping to those types of levels. Is this a three to five-year kind of time range where we would expect most of those to be in that kind of revenue level? How should we think about them?
Realistically, when we put this in this tier 1, tier 2, this is the first time we've introduced that term, and we had some feedback of just how do we get our arms around it. When we look at the total available market that these customers serve, the unit volumes, the test times, is it 100% burn-in or sampling? We can estimate what their buying power is for any given year. Right? The intent of that was that is a range of the kind of customer.
I don't want to get into all the detail, but our 10% customers that we have are forced to talk about, and we're adding one or two more this year. Our 10% and our 10 customers are the likes of Intel and ST and Apple, TI. These are large multinational companies with large-scale applications. For them to be able to do $10 or $12 million in a year, they've all proven that they can do that. Okay? That's the kind of class we're trying to describe that as. We have other customers, including ones that we've announced, and I want to be careful not to not mention them, but I don't want any of my customers to think I don't love them as much. The likelihood of them doing $10 million is nowhere near the same.
We would expect them to do $1 million, $2 million, or $3 million a year or so. As always is the case, at least early on, we can see even customers being lumpy, like maybe they have a good year and then a soft year the next year. If you start to look at these new devices, silicon photonics and Silicon Carbide and some of these others, they're actually growing so fast, it's reasonable for them to be sustainable year after year. Kind of ignoring some of the craziness going on in the world right now, we would've expected this to be an up year even over last year's guidance. Again, I don't think you should just linearly add them all up, but it's not crazy that they all have good years in the same year.
I think it's really going to be maybe two-thirds of them, maybe any given time, are having good years, and then they kind of alternate over time. The goal for us would be to get 15, 20 customers, which we think is reasonable, so that we don't have to wake up at the beginning of the year and think we're totally dependent upon one customer to do a significant amount of our sales.
That's great. Very helpful. That's all for me. Thanks, guys.
Thank you.
Thanks, Tyler.
Thank you. We'll take our next question from Jeffrey Bernstein with Cowen.
Yeah. I'm sorry. I think you cut out for a minute. At one point, you were talking about some business that is not in the guidance. Is that correct? Can you just go back over that?
Sure. Okay. Did you hear me cut out, or did I stumble maybe talking? All right. Okay. Anyhow. What I did, and I'm not going to go back and necessarily read it. I'll just pull it back up again to make sure. This was specifically we've seen in the last several months maybe, and even picking up steam, some renewed activity in our 2D, 3D customer base that have wafer and singulated die systems installed. They actually continually have been buying die packs from us and wafer packs, the consumable, with new product releases. That's been material in business for us. Interestingly, they had not added any system capacity last year.
We've been brought into several new programs that we are aware of now that appear to be a good fit for our products, and we're just getting our arms around it right now. It's very interesting that there's some renewed activity there. If you followed us several years ago, one of the tough years we had was we started the year with a customer saying how great it's going to be. We forecasted that thinking that that was a gimme, and ultimately they pushed out that program and ultimately canceled that specific program. I'll tell you, I'm a little gun shy about beating my chest about these particular applications until they're a little bit more direct in light of their eyes.
I didn't feel bold enough to say that we can meet our guidance without any of these, some of these deals could be some material upside to that or certainly offsetting the other possible issues. That's what I meant to say by that.
Got you. Is there a change going on with regard to 100% burn-in kind of thing versus some sort of statistical sampling and that kind of thing? Or is this really just about brand-new programs?
Well, let me do this. The programs that we're in, which include both 100% and sampling programs, are continuing along those. These are new programs. I.e., we actually don't have all the details. I think we believe that one of them is 100% and one of them is a sampling. We don't have all the details. Yeah.
Got you. That's great. Thanks so much.
Thank you. As a reminder, to ask a question, please press star one. We'll take our next question from John Fichthorn with Dialectic Capital.
Hey, John.
Yeah. Thanks for taking my question. How are you? I apologize because I've been on multiple calls, so I just want to apologize in advance in case you've addressed this. I guess I'm partially confused at the progress you keep making. You introduce new products, you get new customers, and yet I don't see it in top-line guidance expectations, top-line performance. Granted, you've got COVID for this quarter, I understand. Where's the disconnect? You've signed up a ton of new customers in the past. You continue to sign up new customers. You've got existing products, you've got new products. You talked at length about them. Everything seems to be going the right way, and it doesn't seem to be flowing through. I'd just love some help with that basic question.
Well, I think if we step back, as the board and I do sit and try to look at the overall business to try and get our arms around it, we've been putting together three-year plans, looking at what those forecasts are. One of the things I think is fair, just at the minimum, was our first lead silicon photonics customers was buying FOX-XP systems, but they didn't go to production yet. We talked about that several times, maybe almost every single quarter. They, for whatever reasons, had not actually introduced the new products that were moving to the wafer-level burn-in, and then finally have. Qualified it, got it to POR. Actually, if you just look at linear time, that was over a couple of years. That was not what we expected. Ultimately it was, okay, when's this going to happen?
We finally saw that ship to production this year, and they started to buy all the contactors. We knew they couldn't go to production because they didn't have the contactors with them. They had these beautiful systems sitting there that were not being used yet. The reason was, is they were bringing up the systems, and they wanted to ramp quickly. We were able to ship a lot of wafer packs all within a month or two, and that's part of the business model to get them to production. Other customers were kind of a mix. Some that had bought after our lead customer but also didn't get to production till this year. Then we had another customer that was like a six-month period.
The customer that we just shipped to with the order that we announced in Q4 is going to be in production here shortly. Interestingly, on our Silicon Carbide customer, they went from 0%-100% within two months. Actually, Vernon, my VP of Sales, was presenting in our board discussion the collapsing amount of time it's taking for the customers to go from, say, when we first ship it to get to production. The lead time on the sales cycle is still a little bit spotty. Some of those have been shorter and some of them. We've been talking to some customers for a year now, maybe a year and a half. We've had other customers that, okay, their first innovation, 1-2, maybe two quarters later, they already are purchasing.
I think there's this balance of either our own over-optimistic or how fast customers are actually ramping or the deployment. I think to some extent, the deployment of silicon photonics devices in the main end customers. Remember the big, or if you know this or not, we have a couple of folks that follow our stock that really track this market, but it's really interesting to talk to them. The big buyers of this market are Facebook, Amazon, Google. It's the large data centers, okay? They have enormous buying power. One of the things was it appears they now finally are buying silicon photonics devices. Before they were buying the traditional, much more expensive fiber optic transceivers. This is a major inflection point for us because the end customers are now buying in volume, right?
Shifting to the higher performance, higher speed devices as well, all of which is good for us. I believe at this point, it feels more believable that there's more data behind that everybody saying it's going to be great, and now we actually have customers that are ramping and that the end users are pulling. What I will tell you is one of the big things we were surprised by is the seeing push outs of capacity. Well, maybe we weren't surprised, but we saw push outs of capacity for those end-use customers starting in the beginning of this calendar year. Some absolutely customers point specifically at the COVID implications of it, and others are, "What happened? Why did the data center slow down?
Was there a push out? It's something that we're trying to understand, but there's no doubt in my mind data centers aren't going away. The move to more fiber optics, the move to the fiber optics down at the lower levels of the data center because of the bandwidth, all the data is there to support it. I think that's a trend that's happening. I wanted to try and put it in perspective.
All those trends make sense. Everything you said just makes sense, except for the fact that it doesn't necessarily line up with guidance that is necessarily lower than the guidance you gave last year at the same point in time. I just don't understand why $25 million-$28 million relative to $28 million-$31 million or wherever you were a year ago makes sense given you have more customers, you're faster to production, you have more new products. Help. Help me. Why are we not at the inflection point I thought we were at in terms of revenues really inflecting higher? What am I missing? Is it slower lead times?
No, John,
Is it push outs?
Sorry.
Is it coronavirus?
Well, we certainly have stated on the push outs that we have directly seen from customers that they themselves are explaining that and saying that that is turning around, and they're planning to move forward. We have some customers that have, for the first time, actually given us heads up of capacity and forecasts. I think related to guidance and that, certainly in the board discussions, there's a sense of how aggressive or conservative should we be in this environment. We want to make sure that we can do what we say we're going to do. There's obviously certain things that are out of our control, but we believe that this is a comfortable number.
Hey, Gayn. I'd like to add a comment related comparative to the prior year. If you take a look just at wafer level burn-in, even though our overall total revenues only went up 6%, wafer level burn-in went up 39% or $5.3 million compared to last year. Even though we didn't get any system revenues in Q4, and there were the pushouts fully we talked about, we did recognize a 39% increase. One of the items built into the numbers was we had a decrease of $4.1 million, if you take a look at our packaged part revenue and our customer service business that included the upgrades of ABTS systems from prior year. Keep in mind, building into our model in FY 2020, were additional revenues in those areas as well.
My last comment is just that, I really think you guys should take a serious look at your board ahead of this year's proxy season. It looks like it could use a refresh under ISS standards. I think that's something that could maybe help you guys progress more rapidly or bring a set of fresh eyes to the entire situation. I would recommend that as a shareholder and otherwise. Thank you.
All right, John, appreciate it. Just to comment, just in general, I know there was a specific question there. Actually very happy with. There's been a transition in our board over the last couple of years of adding some additional board members. Laura Oliphant, who had been a capital director at Intel, very influential, has been wonderful. John came on our board a couple years ago as well. We did actually have our lead board member, independent board member, pass away during the year. There's been some discussions about when's the appropriate time to add a board member, both for any kind of government-related things, and then just honestly with some reasonable balance and expenses and others. I appreciate it, John.
Thank you. We'll take our next question from Geoff Scott with Scott Asset Management.
Hey, Geoff.
Hi, Gayn. How are you?
Okay.
First question. You added a new sales rep, I guess, German-based to cover the European accounts. How long do you think it will take to get some sense as to whether or not that is working out? Precisely what kind of metrics will you be looking at to determine whether or not it was a good move? Is that number of new customers? Is it dollar volume that they're able to generate? Please help me out.
Okay. Well, a couple things there. Certainly, early on, I did and so did Vernon, made several trips and had customer calls with them, and I was really pleased with, I guess, the number of quality customers and the levels within those customers they were able to get to. The last time we were there was, I think, February or March. One of the things that definitely has impacted us has been some of the international travel. I think Ken kind of alluded to it, the expense controls. It's not very hard to have expense controls right now. We've been doing some pretty creative things with our reps and remember, we have employees around the world to ensure that it's not limiting our capabilities for installations and support of customers.
HCT, who's out of Germany and covers Northern Europe, has actually been a good start. I do expect them to move from great customer calls to orders, both in terms of volume, quality, and eventually dollars as well. Those are expectations that we would have starting this year. As well as reps in Southern Europe, and we are working to close on a rep or two in Japan right now. We've already been in discussions, but we didn't formalize anything with them, and looking at some other things. There's absolutely people that are really good at some of these non-silicon application customers around Silicon Carbide in particular, but also silicon photonics that we're trying to bring on. I'm hoping and expecting, and Vernon has metrics around getting more sales through those reps as well.
Okay. Next question. You have been working on engineering for a more fully automated FOX system. Where are you on that engineering progress?
That is a system that is still in development. We haven't talked and given customers exactly what the integration and lead time is due to some specific competitive and other reasons, but that is in process. I'm just going to leave it at that for now.
Okay. Last question. I'll have to go back and kind of parse the statements made about the tier 1 customers, tier 2 customers, things like that. My gut instinct is that if you add the revenue guidance for fiscal 2021, that virtually all of that is coming from further penetration of existing customers and no need for any real new customers. Is that fair?
That is fair, I also want to just point out that's not because we think we aren't going to win new customers. If you want to take that from how conservative we are, it's like, guys, I'm not starting off this year thinking that if we can just win that one big guy, I can make my numbers. That is not the case. We can build up that guidance entirely with current customers and applications.
What my sense was any new customer would represent an upside from that guide range that you gave us. Is that good?
Yes. That is true. To be balanced, it obviously creates some potential for de-risking, but yes.
Yes. Okay. The guidance is really going into production from existing customers as opposed to a hope list that you're going to sign up any large new customers.
That's fair.
Okay. That's all I have, Gayn. Thanks.
Thanks.
Thank you. We'll take our next question from Tom Diffely with D.A. Davidson.
Hello. Yeah, quick question on the balance sheet. Assuming that you do get a couple of customers that happen to come at the same time, what can you say in terms of working capital or inventory builds and the balance sheet?
All right. In terms of the balance sheet and inventory builds and capitalization, we have $8 million in inventory as of the end of May 31st. 6.5 of that is in our FOX-P product, and included in that is about $1.7 million, what we call our demo lab, that is turnkey. It allows for quick turns on that. The benefit of that is we can turn quickly, and we do not need to add inventory. We're in a position where we don't expect a tremendous increase in inventory to meet our goals during the period, and don't expect a whole lot of additional capitalization to be able to fund that.
Tom also, just to put it in perspective, we had, and it's interesting, I think I talked about it the last quarter and didn't do it this, I certainly had enough content already. This last year, one of the other big things that we did, we just did our end-of-year plan, is we significantly lowered our lead times for our wafer packs and increased capacity. We're able to do short-term quick turns on fully custom wafer packs and die packs. What's important about that is we don't need any material around for it. Given that half of our business is die packs and wafer packs, you could say, "Oh, wow, that's it." We really, in most cases, we're being paid by our customers by the time we pay our vendors on those things.
Lastly, actually, we didn't talk about it because we also didn't use it. We did put in place a line of credit this last quarter. I don't know, you want to cover that a little bit, Ken?
Sure. It actually was prior to last quarter, but with our banker, Silicon Valley Bank, we have a $4 million line of credit that's available to borrow for domestic AR. We have not borrowed any of those funds that are available to us through the bank. We have that ability. Plus, we also have customers that also allow to factor any open AR that's actually at a rate that approximate prime, so it's very favorable with these tier 1 customers, in fact, one of our lead customers. There's many avenues that we have to really build our cash position as needed.
Okay. It sounds like
Sorry, go ahead, Tom.
I was just going to say, it sounds like even to exceed the high end of your guided range for the year, you've got plenty of working capital inventory and manufacturing capacity already in place.
Well, that would certainly be our goal. I want to be careful of getting too far ahead of my skis on this, given kind of where we're at. Yes. Tom, one thing, and I think you know for those folks that have followed us for a while, I haven't said the words in a while, but we still, all of our standard quotes are 30% down payment, Tom. Okay?
Okay.
It's typically quite rare that we look the other way on some of those things, and in many cases, with contractual obligations as part of these. We get large system orders from these tier 1 customers for either wafer packs, die packs, or systems. They're actually giving us down payment, which in many cases is the majority of the actual material cost of the sale itself.
Okay.
They're not cancelable either. From the old classic, "Well, what happens if you get a $30 million order? How are you going to afford it?" That has been removed from our vocabulary.
Well, I appreciate all the detail on the call today. Thanks.
Thank you.
Thank you. We'll take our next question from Larry Klabunde with Solina Capital.
Real quick question on your Silicon Carbide application, the wafer level system that you've sold. That does both the ATE test and the burn and reliability test in one fell swoop, doesn't it? You don't need to do an ATE on the wafers after it comes through the fab with your system?
Well, it depends. There are certain test steps that they functionally test the devices with our system. One of the very unique things that they're very pleased about, in fact, there are white papers out there. I'm not a big message board, but I did see that somebody posted on the Yahoo message board. They have done white papers on the street talking about, and beating their chest about their new wafer level burn-in system. They did specific comparisons, talking about the 90,000 devices per week capacity compared to the 7,000 on their system, plus the ability to test it before it gets packaged into these modules. On top of it, they specifically point out that one of the key things is they can discern good from bad devices.
As the devices fail during the burn-in process, they can see exactly where they're failing, which is very unique for a burn-in system. The testing capability gets into voltages and currents, and opens and shorts and things. We can give them 100% certainty that it's validly tested, and that the proper voltages and currents were put there. That's a big deal. There are other tests that people do in an ATE insertion before and after burn-in typically, but in the case that we fail a part, they won't even test it. They don't think it's different. It's a little of both.
You're saying that the throughput on the wafer level is something like 12 times what the part burn-in system is. I think you said in your comments that because you're catching those failures, the infant mortality failures in the die level on the wafer level, you haven't put any more money into it, or they haven't put in packaging and so on. Did you say it was up to five times more expensive once it's packaged when they find that failure and throw it away?
The numbers I said is 4 to 10x or so. The specific things, if you go look at I actually have one in my hand right now, okay? Too bad you guys can't see it. I'm physically holding one of these high power modules that goes into a car. It's this large metal frame with a large heat sink on it. It's encapsulated. This thing, they put 8 or 10 of these MOSFETs in it in parallel, and then they'll sell it. If you go out on Digi-Key or anything else, you can find it from the folks. There's several people that are out there. I don't want to go into which one is my customer, although they will show up as a 10% with our 10-K, right? These things are doing hundreds of amps.
They're put into this thing, integrated in, encapsulated, and then they burn this module in, and then one of the devices fails, and it's pretty high failure rate. They're throwing away 10% or more, I don't need to give you the numbers, but that's an easy number, of this entire module. The package itself might cost $100, $200 to build. The die inside them might be in the fives of dollars a piece or tens of dollars a piece, because that's the price of them when they're discrete. That's a reasonable thing. When they throw that away, they throw the whole module away. It's a big deal. That's why everyone's scrambling. They all have edicts get to wafer level.
The throughput's like 12 times higher, give or take. You're catching those failures at the lowest cost before any more money's put into them, which sounds like a huge savings.
Yep
The roadmap down the road is to go to 8-inch wafers, your system's capable of 12-inch, so it's future proof. Of course, the economics getting even better in that scenario. I guess I go back, why isn't everybody going to this? Is everyone currently burning in their components in the parts system? Why isn't it such a no-brainer to go to your wafer level?
We have a large punch list, customer list of the ones that we are targeting. A reasonable majority of them we have not been in contact with. That's just the truth. Others we have been, and we started to make engagements, and we're going to talk, starting at the top of the list, if you will. Now that we have proven the feasibility and the capabilities, we are engaged with multiple other customers on this. There are certain tests that we can do, and there are certain tests right now that we do not yet have proven on our FOX systems that we are also working on. We're working with one or more of the largest Silicon Carbide customers on working through that application. I think that this is an area that maybe how do we get in front of customers?
How do we make phone calls? How do we get it? That's one of the reasons to kind of beef up and make the changes in our sales so we can get in front of more customers.
Right. It seems like this approach is a no-brainer, but assuming the industry would go this route, is there an idea how many systems could potentially be involved for the Silicon Carbide application? How many XPs?
Well, I think I specifically walked through some of the math there. I talked about 500,000 wafer starts a year.
Right.
I referred to the burn-in times as much as days. If you kind of walk through the math there. There are multiple hundreds of wafers worth of capacity out there, we believe today, and growing significantly. What I mean by wafers of capacity would be how many blades, in our case, need to be installed worldwide. That market is substantially larger and, to be blunt, the whole world only has one XP in production right now. That's the first one. That is why it's so exciting. That customer is going to be taking on more. We are going to win other customers. It's sort of a mad scramble right now.
It seems like if you get this.
And I-
No. Go ahead.
I know this may sound interesting, but anybody who also follows some of those customers, we actually have talked to a couple of the CEOs. We'd like to talk to them all.
You're an enabling technology on silicon photonics, which is an emerging new technology that has great potential. Hopefully, you'll be recognized as an enabling technology provider for Silicon Carbide, which looks like it has tremendous potential, but one or two more proof points.
We think Silicon Carbide is and will grow. It's already larger than silicon photonics and will grow much larger than that.
Mm-hmm. All right. That's all I had. Thanks for your time.
Thanks, Larry.
Thank you. We'll take our next question from Charlie Doe, private investor.
Can you break down the revenue for the lower end of the guidance between product sales, consumables, and service? As an alternative, can you provide the level of product sales you think will be needed for the lower end of guidance for
Let me give you some ranges that are reasonable based on historical. We historically are recording somewhere $800,000-$1 million a quarter in service support. Those are the service contracts, time and material things. Using that going forward makes sense. Take $4 million out of that and you got service support. Generally, we've been tracking maybe 30%-50% of consumables as a normal run rate, and you could use that as a reasonable. I think we do believe we'll get some package part business this year, but probably not a significant number. The rest will all be between FOX and the FOX consumables. We do think our package part business is likely to pick up the following year, but I don't want to get out there too far.
From a margin and mix perspective, our FOX systems and wafer packs have a very good margin and are what we call kind of a material margin, which is the price minus the raw material or the direct manufacturing costs are in the 60%-65% margins. Whether they're the consumables or the systems are about the same from a mix perspective, and it's actually somewhat similar for the service support now. Our overall, we're probably going to be less subject to mixes, although there are certain configurations and certain consumables that have higher margins.
Great. Do you expect any material changes in the R&D expenses for fiscal year 2021?
We generally run about $1 million a quarter. I think we've done a little less than that in the last because of just some R&D spend with respect to where we're at in the programs. It's my expectation that that's about what our run rate should be.
what can you tell me about-
Right, Ken?
Yeah, you're actually correct. We've been running about 800,000 and plus the last several quarters. We are going to be ramping up. Keep in mind, we have a few of our programs that will be ramping up.
Yes. That they did.
third and fourth fiscal quarters.
Ken and I are not in the same room, but we have a audio Zoom on our computer, so we can each sort of signal to each other.
What can you tell me about?
Sorry, go ahead.
the prospects for new products, their timing, and the potential markets, given that you're going to continue with that R&D spend rate?
We specifically already either alluded or directly stated that we are working on a new platform product in the package part that has very high voltage on it. That looks actually really exciting. Customer feedback has been very positive on it, and we think that it opens up. There are certain applications for Silicon Carbide, for example. I mean, a discrete MOSFET. Customers may want to put it only in a package part burn-in system. There's just not the capacity out there to address it. That's an area that we've had other high voltage from 300 volts to 600 volts, and then the SiC and gallium nitrides are up to 1,600, 1,700 plus volts. That's an area that we see is going to drive some things, that we've got R&D spend going on.
We have some automation programs and some other things that we're working on that we believe not only enhance our ability to address equipment customers, but also some high-volume market opportunities. I have just discussed briefly in our calls that we are talking with some of the memory suppliers related to their roadmaps for wafer level burn-in, and we'll continue to provide information as we get further along in those discussions.
Thanks. Appreciate the input.
Thank you.
Thank you. We'll take our final question. It comes from Jeffrey Bernstein of Cowen.
Yeah. Hi,
Hi, Jeff.
Two quick ones. You just touched on gallium nitride, and my question was, what's going on in terms of opportunity for you in gallium nitride RF and gallium nitride power, and what have engagements kind of looked like so far? I have one more.
We might actually already be testing a gallium nitride wafer. Apologize. I know that that is the plan, but it is a power device.
Okay.
That's interesting. Hey, operator. I got one of our people that was just on the call calling myself. Did we have a problem? Is everybody still on? Okay. Anyhow, they absolutely are doing both Silicon Carbide and gallium nitride. To be blunt, they look the same to us. They both have similar reliability issues. You got to have a certain, I don't know if I'd say sense of humor, but if you're a tester guy and you're testing semiconductors, you love devices that fail. Okay? I mean, that's a good thing for us.
Right.
Fail in an application where it matters. The other thing is you want devices that fail and are never going to stop failing. Silicon Carbide is actually a very mature process in many ways. It's just that substrate is inherently not going to give you the crystalline structure and ever get to the defects that silicon will.
Yeah.
People know that.
The compound semis are.
You can't get silicon, but you can't get semiconductors up to 1,700 volts with the efficiency.
Right.
Okay, you do a burn-in step, and all you do is one or two days later, you read it out. These aren't actually very expensive steps even on our tool. From a cost effectiveness, it's cheap. People were just surprised we could do it. For folks that weren't on the call last year, when we first engaged with this initial customer, they were like, "What do you mean you can do this?" We said, "Yeah, we'll show you." We just simply offered them to buy a wafer pack, and then we tested one of their wafers. They didn't even know that it seemed feasible. They didn't think it was at all possible. We tested their wafer. We send it, I think we tested two or three of them, sent it back. It went through their entire qualification process.
We not only told them which devices failed, but what time they failed. They went through the whole qual process. It caught 100% of the failures. They then turned around and said, "Can we ship you more wafers? And here's a PO." We actually, while we were building our system, continued to ship wafers for them. They said, "Can we please borrow more time?" We actually shipped out of our apps lab. We were shipping them wafers. I believe they were shipping to customers because they were like, "This is fantastic." That's why they immediately ramped. That's been a very positive experience for all of us.
GaN RF, is that a different test structure?
Kind of like
to do that or?
Far, I have not heard about people driving the RF part of it, these tiny little transceivers, for the same level of reliability, but our ears are open. I think it's just that maybe the criticalness of the first one, the first application was these power modules going into automotive where they were stacked. I believe that people will be doing this for discrete, and it will not surprise me if we go to the RF side. The system capability is there.
Yeah. Got you. Okay. Great. Just back on the question on the relative guidance, which kind of jumped out to me as well, but I think what I'm hearing is, in each one of these markets, we've had a Gartner hype cycle. Gallium nitride, Silicon Carbide, 5G, silicon photonics, 3D sensing, AR, autonomous driving, every one of these things. We all know they are all coming. It feels like we're just in that early phase where you just don't have enough that's all going to production and getting out there in the world to really get a trend going for you guys.
Hopefully, that means that this year's guidance has sort of been shaken down to take all that into account, and say, "We're going to stop hoping this stuff hits the inflection point, and we're going to let it hit the inflection point when it does." Is that a proper characterization?
I think majorly so, and that part about there's only so much we can do. When we have a customer that says we're POR, they have the products. I mean, we had specific quotes we were giving to ourselves and the board. It's like, "Nobody else has anything like this, period, at any price." Of course, then it's like, then why can't we charge more for them? That's a different discussion. We know that they're planning to, and again, my intention was not to spend a lot of time talking about COVID. There are some businesses that I think we all can just wonder what the heck is going to happen. This shouldn't be one of them. I mean, our semiconductor companies are all engaged. All their fabs are open. The end-use applications for us are not being impacted.
We believe there's nothing materially impacted related to anything related to COVID or pandemics or international travel or hotels or something that should have any impact. Certainly more communication, more data, all of that stuff is in play. I think that I'm comfortable with where we're at. One of the discussions was, "Should we go out with guidance?" It's like, absolutely, we should. We hope to have a really good year this year.
Great. Thanks very much for the time.
Thank you, Jeff. Operator?
We have no further questions in queue.
Okay. Okay, I appreciate it. Folks, we appreciate your time with us. As always, if you have any questions or follow-up questions, send me, get a note out to us or through MKR, and we can set something up from there. Okay? Thank you all very much, and have a good one. Everyone, stay safe.
This concludes today's call. Thank you for your participation. You may now disconnect.