Good morning, and thank you for standing by, and welcome to the SkyWater Technology third quarter fiscal year 2021 earnings conference call. At this time, all participants are in a listen only mode. After the speaker's presentation, there will be a question-and-answer session. To ask a question during this session, you'll need to press star one on your telephone. Please be advised that today's conference is being recorded, and if you should require any further assistance, please press star zero. I would now like to hand the conference over to your speaker for today, Ms. Heather Davis . Thank you. Please go ahead.
Good morning, and welcome to SkyWater's third quarter fiscal 2021 conference call. With me on the call today from SkyWater are Thomas Sonderman, President and Chief Executive Officer, and Steve Manko, Chief Financial Officer. I'd like to remind you that our call is being webcast live on SkyWater's Investor Relations website at ir.skywatertechnology.com. The webcast will be available for replay shortly after the call concludes. During the call, any statements made about our future financial results and business are forward-looking statements. These forward-looking statements are subject to risks and uncertainties that could cause our actual results to differ materially. For a discussion of these risks and uncertainties, please refer to our filings with the Securities and Exchange Commission, including our earnings release filed on Form 8-K yesterday and our perspective filed April 22nd, 2021.
All forward-looking statements are made as of today, and we assume no obligation to update any such statements. During this call, we will discuss non-GAAP financial measures. You can find a reconciliation of these non-GAAP financial measures to GAAP financial measures in our earnings release, which is available on our investor relations website. Unless noted, all comparables referenced today are versus the prior year or third quarter of fiscal 2020. With that, let me turn the call over to Tom.
Thank you, Heather, and good morning to everyone on the call. I've often spoken of and will continue to emphasize the historic nature of the post-Moore's Law technology paradigm shifts that are occurring right now, as well as the role SkyWater is playing to enable and drive this change. Every day, we are working with our customers to co-create solutions through our highly differentiated Technology as a Service model in areas such as rad-hard microelectronics, photonics, MEMS, superconducting, imaging, microfluidics, and advanced packaging. The disruptive technologies that we enable support megatrends across the economy that are shaping our world, like next-generation communication networks, advanced and quantum computing, autonomous vehicles, electrification, renewable energy, rapid bio diagnostics, genetic sequencing, smart biomedical devices, and state-of-the-art defense systems.
Interestingly, in addition to technology and market-driven paradigm shifts, the supply chain challenges across industries, which you are no doubt hearing about, highlight the need for our nation to enhance domestic infrastructure. As the country is coalescing around the concept of semiconductor sovereignty, SkyWater plays an increasingly critical role in supporting the vision of reestablishing the U.S. as a technology manufacturing leader. I share all this to highlight our collective excitement about SkyWater's opportunity to address major needs in the industry for disruptive technologies, the unique infrastructure these technologies require to scale, and to make SkyWater and the U.S. leaders in these emerging categories at the same time.
In our call today, I will cover our high-level Q3 financial results, do a deep dive into our Advanced Technology Services business, highlight several exciting areas for expected long-term growth, and then spend some time unpacking some of the challenges we see in our business today. In the third quarter, total revenue grew to $35 million compared to last year, led by increased Wafer Services revenue, which offset a decline in Advanced Technology Services revenue. Supply constraints and delays in expected U.S. government funding had a near-term effect on the company and our customers, resulting in a revenue recognition delay for certain programs. We incurred higher cost of goods as we ramped up hiring and employee retention efforts to support our demand outlook. In addition, we continued investments in building our radiation-hardened and advanced packaging capabilities.
These investments outpaced revenue growth in the quarter and were the primary drivers of gross margin of -5.2% and negative adjusted EBITDA of $2.7 million in Q3. We continue to aggressively ramp our Advanced Technology Services and Wafer Services businesses and made important progress on our rad-hard and power management platform qualifications. We also completed the transfer of two ATS programs two Wafer Services last quarter, with another five underway. In the last 12 months, we have added 25 ATS customer programs and have substantially grown our sales pipeline. All of this gives us a strong conviction in our long-term model. As the past two quarters demonstrate, our business can fluctuate on a quarter-to-quarter basis. Due to a confluence of factors, our third quarter results were below our expectations.
We underestimated some of these factors, which are the consequences of our rapid growth, but we've learned from them and believe that this will make us stronger in the long term. In addition to what we believe are favorable market conditions for semiconductors today, we believe SkyWater is well aligned to the long-term growth trends that demand services offered through our unique engagement model for the development and production of new technologies. Our approach to the market is called Technology as a Service or TaaS. This business model combines traditional volume manufacturing services, or what we call Wafer Services, with Advanced Technology Services, which is an embedded R&D capability that enables us to monetize this competency. In the TaaS model, customers come to SkyWater with an idea, and we develop the manufacturing process together.
In addition, we are a volume manufacturer, which means customers also get the benefit of high yields and automotive level quality when their product goes into production, as well as the IP security benefits of performing development and manufacturing onshore. We have honed an exceptional proficiency creating novel and disruptive technology with our customers inside a production environment. This unique combination of Advanced Technology Services and Wafer Services inside a single operation is one of our key differentiators and facilitates customer engagement throughout the development cycle, where we demonstrate and optimize their products while refining our capabilities. This symbiotic relationship allows customers to create unique offerings and allows SkyWater to serve customers with differentiated technologies not easily produced elsewhere. The ability to handle development projects from R&D to volume production gives us a lot of stickiness with our customers.
When faced with the requirements of establishing an operation capable of producing their device, customers typically decide that investing in their own facilities or engineering teams is not economical and simply takes too long to capitalize on the market opportunity. This is where SkyWater's value maps to customer needs again and again to help them innovate, drive customization, and accelerate their time to market. We make it easy for customers to engage with us at any point in the technology and product development cycle, but our ATS projects typically initiate during concept and feasibility, technology demonstration, or process development. Ultimately, a successful ATS program will conclude with customers moving into Wafer Services, which is when they scale their product to volume production and do so with a process flow designed to run at SkyWater.
In the ATS phase of a customer engagement, we focus on margin productivity, targeting a high margin. These are all R&D programs centered around cycles of learning and validating technology. As customers move into Wafer Services, the margin per activity decreases as volume grows, but also requires substantially less engineering support, enabling those resources to focus on the next ATS program. The other key component of our TaaS model is that it lowers our capital intensity. Customers invest in their development not just for non-recurring engineering, but also for tool acquisition. If there's a program that requires tooling we do not possess, our customers typically fund the acquisition of the tool. It's not SkyWater's practice to offer exclusivity, which means that we are able to use that tool not only for the lead customer, but also for other development and volume manufacturing programs.
This approach enables the necessary development and subsequent production required to monetize the IP co-developed with our customers with low capital intensity and strong mutual benefit. The positive momentum we've generated with the TaaS model continues to grow, and we believe strongly in our highly successful approach. This is evidenced by our eight ATS wins during third quarter, bringing our total number of active customer engagements to 55, an increase from the 35 we disclosed during our IPO. It's important to recognize that it can take customers 24-48 months to move through the ATS funnel into volume production. Revenue generation from these programs typically accelerates as the programs mature. This year, we've begun the transition of several customers into Wafer Services. These new programs deliver higher margins than our legacy Wafer Services business. The biomedical space is one of four important growth areas for SkyWater.
Multiple bio health programs are moving quickly at our company. Our highly collaborative engagement model is well aligned to needs in this space as conventional microelectronics and microfabrication technologies are rapidly merging with screening and intervention technologies. This ability to co-create across fields, along with SkyWater's ISO 13485 medical certification, brings a strong value proposition from development through production for new capabilities in the areas of rapid diagnostics, genetic sequencing, and a wide range of biomedical devices. We remain encouraged by the pipeline for ATS projects in the biomedical space. We achieved an important milestone with our radiation-hard technology platform fabricating in Q3, the first fully integrated test wafers with copper back end of line processing. Domestic production of rad-hard wafers is significant for the U.S. government in both space-based and defense applications.
This work is progressing and ongoing, and we continue preparations to serve this critical high-value market. For our Florida advanced packaging facility, we recently announced an agreement with Deca related to their second-generation M-series fan-out wafer-level packaging technology. We are working with Deca to become the first pure-play foundry to offer this competitive AP technology solution in the United States. The Gen 2 Deca technology is state-of-the-art in its ability to reduce die pitch and fan-out wafer-level packaging integration schemes. We previously announced that two SkyWater Minnesota customers have selected our Florida facility for their advanced packaging needs. We continue to see increasing momentum and customer interest for our unique AP capabilities, and I'm very excited about the future of our Florida facility. The fourth growth platform for SkyWater is high-performance power management and connectivity.
We have partnered with several customers in this category as we work to transition them from ATS into Wafer Services. One of these customers is working with us to commercialize a new innovative fin-based MOSFET architecture that will provide substantial benefit to high-speed power switching applications. This partner is currently working with their end customer on system qualification, and we anticipate that the Wafer Services ramp for this platform will begin with revenue generation in 2022. We believe the strategic and national importance of improving domestic manufacturing of semiconductors cannot be overstated. The CHIPS Act received bipartisan support in the Senate, and we remain confident that it will ultimately become law. SkyWater continues to engage with multiple state governments to secure the matching funding necessary to capitalize on our nation's strategic investment in semiconductor research and manufacturing.
Last month, we were honored to have Minnesota Governor Tim Walz tour our facility and discuss several opportunities we see to quickly increase the output and capabilities of our Minnesota fab. We believe SkyWater is an excellent example of how public-private partnerships can collaborate to deliver domestically produced semiconductors for government and commercial customers. As the U.S. looks to build infrastructure in this area, we are optimistic that there are various scenarios for which we are uniquely positioned. We intend to be a leader in building our nation's manufacturing infrastructure for technologies that will be major industry drivers in the years to come. In the near term, we expect to continue to face challenges with supply chain and labor constraints. As highlighted, there's ample opportunity for long-term revenue growth for SkyWater to deliver increased shareholder value. Skilled labor is a key pillar of our growth.
Starting in the second quarter of 2021, we began hiring for a range of roles key to scaling the output in our Minnesota fab in response to strong customer demand. The market for skilled labor is highly competitive in general, but the current market is as challenging as I have seen in my career. As we work through this, we are also focused on employee retention for all roles and view it as critical to our future. We're significantly expanding our learning and development capabilities to reduce the time required to achieve high levels of productivity while creating a foundation for all employees to reach their maximum potential. SkyWater was not immune to the challenges in the supply chain for semiconductors this quarter. We talked about increased lead time for tools in our last call.
In the third quarter, both SkyWater and our customers experienced increased challenges sourcing spare parts, substrates, and the chemicals used in production, which did impact our wafer output and ATS activities in the quarter. We have now entered into multiple long-term agreements to secure supply into the future. Q3 revenue was also impacted by a significant complex multi-year ATS program that is now expected to be completed in the first quarter of next year, moving revenue recognition from 2021 into early 2022. These large-scale programs have multiple dependencies, and it's sometimes challenging to forecast the exact timing of these R&D-related milestone achievements. I'll stress that this is a movement of revenue to the right in the model. In 2021, we hired three senior operational leaders with many years of foundry experience.
They are already making a strong impact, driving important performance improvements as we ramp our two fabs. Output was constrained by the ongoing labor and supply chain challenges. Even so, wafer outs for the third quarter in our Minnesota fab increased nearly 60% from the prior year. We were able to achieve this increase despite moving in new tools, improving infrastructure, and onboarding new hires. These challenges are real but well understood in our organization. We are a rapidly growing company, ramping a complex business in a dynamic macro environment. These stressors continue to forge our teams and processes as we scale and optimize our operations to deliver long-term growth and increase shareholder value. I am very proud of all of our employees and what we are doing every day to co-create technology solutions with our growing list of customers.
I will now turn the call over to Steve for information on our financial performance in our recently completed quarter. Steve?
Thank you, Tom. As we build our business and make long-term investments to deliver outsized growth, our financial results are anticipated to be choppy quarter- to- quarter. Total revenue for the third quarter of 2021 was $35 million, an increase of 6% compared to the third quarter of 2020. Advanced Technology Services revenue declined 8% to $22.4 million, and Wafer Services revenues increased 44% to $12.7 million. ATS revenue was impacted by the customer program that is being reconstructed and expected to recommence in 2022. This contract contributed $4.3 million in third quarter 2020 and $23 million in fiscal 2020.
Excluding this program from third quarter 2020 results, we would have shown ATS growth in the third quarter of 2021. We have a large multi-year ATS R&D program originally scheduled to be completed in the fourth quarter of 2021 that is pushing out into early 2022. This moves revenue recognition from the third and fourth quarters of 2021 into early 2022. Customer funded tool revenue, which is included in our ATS revenue in the third quarter of 2021, was $300,000, roughly flat to the $400,000 in the third quarter last year. Wafer output for the quarter increased by nearly 60% compared to the prior third quarter, and Wafer Services revenue increased 44% to $12.7 million. Cost of revenue was $36.9 million, an increase of 43% year-over-year.
Gross loss was $1.8 million, decreasing from the gross profit of $7.3 million in the third quarter last year. Gross margin of -5.2% declined versus the prior year of 22.1%. Non-GAAP gross loss was $0.5 million compared to gross profit of $7.4 million in the third quarter last year. Non-GAAP gross margin was -1.4% and 22.4% respectively. Both GAAP and non-GAAP gross profit and margin declined due to increased cost of revenue. Cost of revenue increases were driven by three primary factors: volume, labor, and long-term investments. The increased Wafer Services output of approximately 60% in the quarter drove increased serving material costs and other variable costs.
At our Minnesota fab, we've continued to hire fab technicians and maintenance engineers to support our expanding volume outlook heading into 2022. We continue to make investments for the long-term growth of the company by building out our rad-hard and advanced packaging capabilities. Both programs are expected to be long-term growth drivers but are near-term headwinds to profitability. In the third quarter of 2021, depreciation related to the rad-hard program was $2 million, and we incurred $2.9 million in cost of revenue for Florida. Year-to-date, these investments in rad-hard and advanced packaging were $10.9 million combined in cost of revenue.
R&D in the third quarter was $2.3 million compared to $1.1 million in third quarter 2020, as we added executive leadership, engineers to support our ATS growth, and expanded design enablement capabilities to accelerate and support the development of our technology platforms. We are enhancing process design kits to continue building out our technology platforms and completed our S130 PDK in Q3. We're also investing in R&D to bring power MOSFET to market, which is expected to occur in the fourth quarter of 2021. SG&A was $9.6 million compared to $5.8 million in the third quarter last year. The increase was driven primarily by public company costs and stock-based compensation. Excluding non-cash costs of $2.1 million for stock-based compensation, SG&A was $7.5 million in the third quarter of 2021.
This is lower than our previous expectation as we have not met our internal plans for the year and reversed most of the annual bonus accruals for 2021. Adjusted EBITDA was a loss of $2.7 million, declining from a positive $5.3 million last year, reflecting the decrease in gross profit flow through this quarter. Cash used in operations during Q3 was $6.4 million. We invested $17 million in CapEx this quarter on fab improvements aimed at increases in capacity and efficiency. We ended the quarter with $8.4 million in cash and cash equivalents. We paid down $30.3 million on our revolver in the third quarter. As a result of this pay down, total debt outstanding was $35.6 million as of October 3rd, 2021, and we had $60.6 million available on our $65 million revolver.
Total inventory at the end of Q3 was $30.9 million compared to $27.2 million at the end of fiscal year 2020. The temperature differential sensing wafers we discussed in Q2 remain in inventory at $13.4 million. As you update your SkyWater models, the following is some additional color for our expected operating costs for the fourth quarter of 2021. Research and development expenses are anticipated in the $2-$2.5 million range. SG&A expenses are expected to be approximately $9.5 million, including stock-based compensation, and we anticipate annual stock-based compensation to be approximately $13 million. As Tom highlighted, the current headwinds resulted in the delay of approximately $15 million of revenue recognition from third quarter 2021 and fourth quarter 2021 into 2022.
Our business with new and existing customers is growing in both ATS and Wafer Services, and we are continuing to transition ATS programs to Wafer Services. This gives us confidence in our long-term top line growth model of 25%. With that, I'll turn the call back to Heather, and we'll hear your questions on SkyWater.
Thank you, Steve. Please visit the Investor Relations section of our website for upcoming investor presentations. Operator, please open the line for questions.
Ladies and gentlemen, just a reminder at this time, if you would like to ask a question, please press star and the number one on your telephone keypad. Your first question comes from the line of Mark Lipacis with Jefferies.
Hi, guys. Thanks for taking my questions. So, the first one, Steve, if you could just play back the additions to your fixed cost structure and COGS. It looks like on a year-over-year basis, a similar revenue level has translated to gross margins going from, you know, high teens, like a 20% range to flat. So could you. It sounds like I got $2 million of depreciation expenses added into COGS, and it sounds like there was some labor costs also that are going into COGS that raise the fixed cost. If you could just, you know, break that down one more time and just help us understand fixed cost basis now in COGS.
Sure. Happy to do so. Good morning, Mark. It's very important to understand the cost structure of the business because what you'll see in there, we continue to remind of the investments we're making that are flowing through our cost of revenue right now for the long-term growth of the company. We really highlighted three things. It was volume, labor, and long-term investments that are driving the increase you're seeing in cost of revenue. Given that we had a 60% greater output comparing third quarter 2021 to third quarter of 2020, obviously there would be higher costs of personnel, starting materials, and other variable costs that go with that higher output.
Secondly, as I mentioned, we're adding fab technicians and equipment maintenance personnel as we get to our target utilization for 2021, and we wanna be well prepared for the forecasts we have for 2022 and the target utilization for 2022. As I mentioned, long-term investments. In the quarter, we had $4.9 million of costs go through related to the rad-hard technology as well as the advanced packaging platform. This does add to our cost structure that wasn't there over the course of 2020. On a year-to-date basis, we had $10.9 million of additional costs come through that I would call investments in rad-hard 90 and the advanced packaging platform for the long-term growth of our company. Our cost structure did change primarily related to those investments we're making for rad-hard in advanced packaging for the long term.
On a quarterly basis, we have $2 million higher depreciation expense, right?
That's right.
In COGS. Okay. Then can you give us a sense of, you know, I appreciate that you know, you're hiring more fab techs and whatnot. Can you give us a sense of, you know. I guess I'm just trying to get to the variance here, like the cost structure has increased $2 million on depreciation. There's labor that's gone up, and then there's the rad-hard investments. Is this something that now is a something. Is it are these costs that you capitalize and now are amortizing? So is that rad-hard element, is that now a permanent part of the cost structure, and we don't get to absorb those from rad-hard until you start shipping from revenues there?
Is that kind of the idea, or was this like a one-time event on that input to the COGS?
You're correct on the rad-hard. This will be part of our cost structure going forward, and it will be part of the cost of revenue that will be coming through and actually started coming through from a fully baked basis in the second quarter of this year. This will remain part of the investment we're making, the investment we're recurring. Again, rad-hard is very important for the long-term growth of our company, but we won't really start generating any significant revenues from that rad-hard platform until late 2023 at the earliest, so significant headwinds against us on our margin structure, but really good foundational investment for long-term growth of the company.
Okay. Fair enough. Second question, on the $15 million of 3Q and 4Q revenues that pushed into 1Q, can you give us a sense, you know, again, of the variance here, like, you know, what part of that $15 million was from the supply chain or from the constraints or the delays in the government programs. I just wanna be clear, so is this revenue that you have high conviction hits in Q1, and so is that like a $15 million, like, you know, high conviction part of your 1Q revenues? Thanks.
Sure. On the government side, that was a delay in about $3.5 million of revenue. The rest of that would be allocated over what we'd call supply chain constraints that would impact the efficiency of getting wafers out of the fab on a Wafer Services perspective, as well as the delays in some customer qualifications for Wafer Services and delays, as we mentioned in our large ATS program. There is high conviction for that revenue. It's very defined on what those revenue opportunities are. Again, we are seeing those headwinds from supply constraints remaining, and that's why it was pushed out of Q3 and out of Q4 into 2022. We'll see how we handle and what those headwinds continue to remain on the supply chain and some of the constraints that we're currently dealing with.
That revenue is very well defined and identified. It's just a matter of executing on that given the challenges that we're facing at the macro level.
Okay. I just wanted to make sure that I was clear on that. That's not $15 million that you expect to recognize in 1 Q, but rather some time in 2022.
That's correct.
Gotcha. Thank you.
Your next question comes from the line of Krish Sankar with Cowen and Co.
Hi, thanks for taking my question. I had a couple of them too. Tom or Steve, given the fact that, you know, the cost structure has changed, I'm curious, you know, in the past, you said your long-term gross margin target is 40%, so I'm guessing, is that a realistic expectation or are we gonna reset it to something lower given that the fundamental structure of the business is changing?
Yeah, I still think that's the long-term target that we've established. We still have conviction and belief in that long-term target. Obviously, we're seeing some near-term headwinds, significantly making the investments in the rad-hard and advanced packaging platforms. As I talked about in the last quarter, obviously, as the revenues grow, that will help us to expand our gross margins. We're seeing more of the opportunities for our long-term growth materializing, and that's why we still have conviction in our long-term growth model to 25% top line revenue as well. As we achieve that and realize the return on the investment, the significant investments we've been making over the course of 2020 and 2021, we believe we can achieve those levels in the long term.
Yeah, I'll just add that the thing that we're doing this year that's very important is moving programs out of ATS into Wafer Services. Today a lot of our Wafer Services volume is legacy technologies that we're running for Infineon, for example. As those new technologies ramp, which is underway, and more to come, certainly a major power management platform is coming in, into the fab from a Wafer Services perspective. You're gonna see that gross margin improve on the Wafer Services side, while at the same time remaining strong on the ATS side, especially as we bring in all these new customers. Over time, we absolutely believe that the not only long-term revenue model is going to hold, but also the improvements in gross margin.
We are starting up, again, a new capability here in Minnesota, a new fab in Florida, and those are drags in the near term, but are all enabling capability to drive margin growth in the out years.
Got it. Tom, I just wanted to sort of dig a little deeper on your 25% long-term revenue target. Can you just help us understand, you know, how much of the 25% is coming from ATS, how much from Wafer Services, how much from tool revenues, and how much is really dependent on future government funding? I'm just trying to figure out if you can help quantify the 25% growth into those three to four buckets.
Yeah. I think, you know, ATS will continue to be a strong driver. As I noted, we've continued to bring in a lot of new customers. The business today is 2/3 ATS, 1/3 Wafer Services. Think of ATS again as a high-margin business, but it's funding our pipeline for future Wafer Services business. Having a strong ATS business today indicates long-term Wafer Services growth in the long term. Of course, we have rad-hard. That platform will be coming to market. As Steve alluded to, we're going to complete process technology qualification early next quarter. We will then go into manufacturing qualification and product qualification with end customers, and then system qualification leading into volume ramp in 2023. All that plus our ROIC platform, another very important program we're doing with the government.
These are all long-term growth drivers. Advanced packaging, many of the capabilities we're putting in Florida don't exist today in the United States. We partnered with imec. We've now partnered with Deca. We have another partnership we're working on for DBI. These are all foundational capabilities that we believe are gonna be long-term growth drivers. I would not think of tools. You know, tools, you know, come and go when we engage with the customer, as I alluded to in my remarks that sometimes the tool doesn't exist, so we get the customer to buy the tool. That I would say is an infrequent mechanism. It's really gonna be driving growth in ATS and then the transition of ATS into high-margin Wafer Services, and that's what's gonna enable the model we're talking about.
Got it, Tom. I just wanted a quick follow-up. I understand, you know, your argument about, you know, the CHIPS Act and the usefulness for that, for you and U.S., you know, manufacturing. I'm kind of curious. There's also, you know, [audio distortion] Act that's being floated around where you get a tax subsidy for investing in the U.S. I'm kind of curious, if we go that route, does that help you given the fact that you don't have very high profits or you're not paying a lot in taxes. I'm just trying to figure out, you know, how this government funding, you know, benefits you if it comes in a different form as a tax subsidy versus a direct investment.
Yeah, a great question. Again, a lot of the mechanics of what the CHIPS Act will actually look like are yet to be defined. There's a USICA, which is the broader component tied to innovation investment in addition to manufacturing investment, so that's important to understand as well. SkyWater will make a lot of money off the mere fact that there's gonna be more innovation, more investment going into R&D, which plays very well into our ATS model. But we're also working with three state governments on public-private partnerships. The way the CHIPS funding will work will be the government will supply investment for new manufacturing capacity that will be complemented by state governments, and then industry will also be expected to invest. The goal is to not only innovate in the United States but add scale.
That scale will occur in out years. It's not gonna have an immediate impact. As I alluded to, we're talking with the State of Minnesota.
At both the executive branch as well as the senators and representatives from the U.S. government in terms of how we can accelerate adding capacity into our Minnesota fab so that you can resolve some of these near-term supply constraints. I believe that we have a great long-term strategy tied to CHIPS, tied to USICA. That said, the delays that we're talking about are programs that are already awarded, have nothing to do with whether or not that bill ultimately passes. I think we have a great relationship with the U.S. government.
We're doing things critical to national security and getting these programs reconfigured around new contracts, longer timelines, more tiered pricing, are all things that are gonna benefit SkyWater in the long term, even though they've created some near-term headwinds in terms of them being, you know, funded for our company.
Got it. Thanks, though.
Your next question comes from the line of Harsh Kumar with Piper Sandler.
Yeah. Hey, guys. Maybe I missed it, but did you guys give any idea of what revenues you're expecting in the December quarter?
No, we didn't give any items on that. We did talk about the $15 million that was pushed out from the second half of 2021 into 2022, and so that was the information we gave on some revenue expectations from the forecast we have going forward.
Okay. Can I ask you on the same topic then, should we expect in the absence of the $15 million or any other kind of thing, should we expect on a flattish trends for the December quarter on the top line, or is there any reason for me to believe that there'd be more wafer services or ATS contracts that might come to fruition and sort of range revenues?
Yeah. You can compare the fourth quarter of this year to the fourth quarter of last year for comparison purposes. Remember, in the fourth quarter of 2020, we had an extra week in the year that was coming through, which is about $2.5 million of revenue that came through in the fourth quarter of last year, given that extra week, which is not repeating again in the fourth quarter of 2021 this year.
Okay, great. It might again be down then is the way to look at it, 'cause I think you guys from, if I'm not mistaken, $39 million. Okay, got it. Okay. Fair enough. Fair enough. Maybe Steve, you could talk about the $15 million. What is your visibility like? I think Mark asked earlier about this piece of revenue. Could you maybe talk about the timing and the certainty and the visibility into this? Will it be spread out kinda uniformly into next year, or is this gonna come as one big chunk at some point in time, and do you have visibility into that timing?
Yeah. We definitely have visibility into the changeable nature of what that revenue is. Again, what we don't have the visibility into is the timing of when that comes through. Given, you know, what we were facing right now, we still see some of those constraints coming through in the fourth quarter, and we know that, which is why that revenue was pushed out into 2022. Depending on how we navigate some of the macro impact that we're seeing, as well as some of the ramping and efficiency that we're seeing in the fab, that will clearly impact the timing of when that revenue actually is materialized over the course of 2022.
Got it. Then I had one on cash as well. You guys paid down, you know, quite a bit of debt, leaving your cash balance of $8 million. The question is: How much do you guys need to have on the books to tangibly like run a well-funded operation, and do you think you might be stretching it here, and what was the rationale again for paying down that debt, as a young growing company that might need cash?
Sure. You saw that we paid around $30 million down on that revolver. That's what we did in the third quarter. Again, we can at any point in time draw that amount back over the course that we need that funding, so that's perfectly accessible. At the end of the quarter, we had access to over $60 million on that $65 million revolver to draw upon. We were making significant investments over the course of 2020 and 2021. We talked about $17 million of payments we made for capital expansion that we made in the quarter. We also had some non-recurring expenses of about $10 million that we funded and paid this year, which are a one-time event, so we don't expect to continue over the course of 2022.
On another note, we talk about some of the royalties that we're paying on our ATS revenues. Those cash outflows were approximately $9 million on a year-to-date basis. We're excited about that rate decreasing and dropping by 50% in 2022 when going forward. We're getting a lot of investments in items behind us over the course of 2021 that won't be repeatable over the course of 2022 on cash outflows going forward. We do expect to make draws on our line over the course of the next four quarters, but that amount that you'll see drawn on that line each of those four quarters should decrease as we grow our revenue and start to recognize some of the return on the investments that we made over the course of 2020 and 2021.
Understood, guys. Thank you.
Your last question comes from the line of Raji Gill with Needham & Company.
Yeah. Thanks for taking my questions. I appreciate it. Just to follow up on the $50 million push of revenue. Can you just clarify? You said multiple programs in ATS in which the revenue is being pushed into 2022. I'm trying to understand the piece parts of that. You mentioned that there was a government deal of $3.5 million. Is that an isolated event? Separately, there are other ATS programs that are being pushed out because of supply constraint issues and your inability to you know, qualify products. Could you parse that out clearer? Because it kind of speaks to the idea of whether that $15 million you know, could be recoverable in 2022.
How much of it is, you know, so to speak, in the bag, as you mentioned, Tom, you said it's just moving the revenue to the right of the model. But how much of that is still dependent on qualifying customers, getting access to wafers versus government funding being delayed? There's just a lot of confusion there in terms of how that $15 million is being parsed out. Please specify that in detail. Thank you.
Yeah, I can give you some additional color, and if Tom needs to, he can tack on at the end. If we go to the Wafer Services first, again, that's where we're seeing some of the constraints on really getting the efficiency out of the fab. The orders for those wafers are there, they're concrete. It's a matter of us getting those wafers out. We're hitting some of those constraints on having our tools up and running, when the spare parts need to come in, the availability of those spare parts, as well as the equipment maintenance personnel, hiring, training those personnel, and also executing on our third-party suppliers that provide maintenance services. Again, with everything taking place in the market, our third-party providers are also constrained and not as timely as what they would have been, given the constraints that they're currently facing as well.
Those orders on the Wafer Services side are there. It's a matter of getting efficient, getting our fab up and running to the capacity that and efficiency that we believe it can, and getting those wafers out the door. On the second side, moving over onto the ATS side. ATS is really. It's not a matter of if that revenue comes, it's more of a matter of when. Some of the milestones that we were working on, we've already identified would not be achieved in the fourth quarter of 2021, and those are pushing out into 2022. As we achieve those milestones, that's when that revenue will be recognized.
That was one item that we saw on the milestone push, and then we have another item that is a customer qualification, given the constraints that they're facing as well, that delayed that by approximately a quarter as well. Those are the items that pushed out, you know, the amount of the $15 million in addition to $3.5 million I alluded to related to the government funding.
Yeah, just to add, you know, the power management platform that's going to be coming to market later this quarter, going into next year, not only requires the wafer fabrication, but also the assembly test and then system-level qualification from the end customers. That process is also feeling the effects of some of the supply chain constraints that are being talked about. The ability, obviously, for us to ramp that program and get it moved into Wafer Services in volume is somewhat dictated by getting those final qualifications completed. We remain very confident that it will be a highly differentiated platform, create a lot of disruption, and frankly, we think it'll allow us to reset some of the conditions in the market as it relates to foundry-delivered power management devices.
We're really excited about that program, coming into volume as we enter next year. The original plan was frankly to have it go into volume in Q4, and because of some of these constraints, we just have not been able to get that done.
Okay. Thank you for that insight. Just to clarify again, so there's a $3.5 million government deal that is being pushed because of delays in funding from the government related to that program. That's a separate item. There's a power management platform, which was part of ATS, I guess, supposed to hit into Q4, but it's going into next year, given the same kind of issues about constraints. There's kind of orders on Wafer Services that you're trying to get out of the fab, but because of the constraints and the third- party ramp, it's all kind of combining to having this revenue being pushed out.
If that's a fair kind of statement, then I wanted just to touch base a little bit on the previous questions related to the cost structure. You know, thank you for the insight on the details related to the investment for rad-hard and the advanced packaging. It was $4.9 million this quarter. Year- to- date is $11 million, $10.9 million. You said that's really not gonna accrue those costs until late 2023 at the earliest, I think is what your words. You said there's gonna be significant margin headwinds related to that. How do I think about then the margins on a go-forward basis?
Are there gonna be more investments in rad-hard and in AP, advanced packaging going into next year, kind of layering on top of the existing investments? Because it does appear based on what you're saying is that the cost structure is gonna remain fairly high for some time to come. I just wanna clarify that. Appreciate that. Thank you.
Yeah, happy to clarify that. Those will be headwinds. What you said was accurate, that those are gonna be recurring headwinds and items running through cost of revenue over the course of 2022 and 2023. As far as additional investments go, we have about $29 million or $30 million in capital investment that hasn't been placed in service yet. Once that $29 million is placed in service, typically we depreciate over a seven-year life. Once that $29 million is placed in service, that will obviously have an impact on our depreciation, which will run through cost of revenue as well, in the near term. We'll continue to make investments in advanced packaging in Florida as we continue to build that out.
Both of those programs, whether it's rad-hard or advanced packaging, there's getting the fab up and running, getting the tools qualified, and building out those technology platforms over the next 12-24 months to really allow us to have a foundation for growth going forward. We don't expect that growth and that return to really start to materialize until late 2023.
Got it. Just last question, Tom. You mentioned that there's a 24- 48 month timeframe to move the designs that are in kind of the ATS funnel ultimately into volume production, so moving them to ultimately to the Wafer Services part of the business. You're racking up more ATS wins, which is good. Could you just remind us, you know, as you look forward, you said that revenue accelerates as kind of the product can mature through this funnel. Could you remind us how that process works again? The reason is that, you know, there could be a, you know, there's a mismatch of timing of revenue relative to kind of, you know, the growth.
How much of that growth that you're baking in for 25%, how much of that is baking in, you know, X percentage of customers moving into volume production? Just wanna get a sense of how that process works.
Yeah. I'll outline it, and Steve can add any color. The way we tend to look at it is there's new ATS and there's mature ATS, and then there's new Wafer Services and mature Wafer Services. We have been bringing in a lot of new ATS customers, 25 over the last 12 months. Those typically start out at a relatively small run rate, and then as the programs pick up momentum, the resources get fully deployed, the R&D cycle of learning begins to accelerate, you'll see them move into mature ATS. Again, depending upon the complexity of the program, that can be a two-year window, or it can be a four-year window.
As programs begin to get toward the end of the development cycle in years two-four, they'll be preparing for new Wafer Services or the transition out of development into volume manufacturing. You'll experience a J-curve effect as the R&D slows down and the volume begins to pick up. Over time, what you're gonna see is the customer base grows, and we fill out that entire funnel. Each individual program will have less significant of an impact on our overall model. We certainly expect that the biomed programs that we're doing, they tend to be more in the two-year timeframe to get to Wafer Services, whereas complicated programs like, you know, rad-hard, for example, take a longer time.
The whole idea is to have, you know, customers constantly moving out of ATS into Wafer Services while backfilling the funnel. Again, the unique thing about SkyWater is all these technologies that we're developing are being done on our TaaS platform, which is us becoming the single source provider on new technologies, new platforms, new products that are going to market, all leveraging the unique, you know, capabilities we have. Our customers are partnering with us to fund our development pipeline to bring new competitive products to market versus traditional products that are really the ones you hear about being consumed today.
It's really a new paradigm for the market, Technology as a Service at the same time as the industry's going through a major new paradigm as we bring back semiconductor manufacturing to the U.S. That's creating a lot of demand for resources, a lot of demand for equipment. It's only gonna get more intense with the CHIPS Act passes, but it's all good for our country because we're creating new capabilities that aren't being made anywhere else in the world, and SkyWater is right at the center of that.
That's what gives confidence in our long-term model still. We're seeing customers moving from ATS into Wafer Services and being backfilled by new ATS programs, which is exactly how the model is supposed to work. I wanna be clear, those new programs typically do start out slow, and they're not significantly material in first year contributions. You know, they're new R&D programs, stuff that really hasn't been done before, so it's more a year by year as we go. As we're more successful, we typically see the spend really start increasing in year two or year three, so 24-30 months into an R&D program. As we're going into the mature ATS stage, as Tom mentioned, that's what's giving us better visibility into our long-term revenue as we're seeing these contracts look to be more long-term contracts as we talked about last quarter.
That's giving us better visibility into our ATS revenue and what we need to do to achieve that revenue as well. Good contributions coming through, balancing that pipeline, but it usually takes 24-36 months for those new ATS programs to really materialize and generate significant revenue.
Thank you.
You do have a follow-up question from the line of Krish Sankar with Cowen and Company.
Yeah. Hi, thanks for taking my follow-up. Tom or Steve, I just wanted to check on one thing. You know, in the past, your reluctance to give near-term guidance was driven by the fact that it's very hard to forecast the ATS business. I'm just kind of curious
If A, is that still the case, and B, if that is the case, how should we think about just the Wafer Services heading into December and also into the March quarter, given the fact that you might see some revenue overflow from the $15 million in March, but at the same time, there will be seasonality going against you? Thank you.
Yeah. Over time, and we'll see that we're getting better visibility, as I mentioned. Not only will the ATS side, will the contracts be extending and being longer, which will give us better visibility, we'll also have more confidence in the services we're delivering. Again, with the R&D that we're currently doing, a lot of it is new R&D that hasn't been done before. As we further build out and take these individual programs and turn them into platform technology development, that will give us a good baseline to build off on, and there won't be as much learning going through in the new R&D programs. That will give us better visibility into the ATS programs.
As you can see over the past couple of quarters, there is some lumpiness that comes through with these contracts, especially on the timing of achievement of those milestones. Again, you've seen the output increasing on the Wafer Services side. We talked about that, and it increases year-over-year. There's still more revenue on the table, high demand for our Wafer Services, and we expect to continue to be more optimized and more efficient in driving more wafers out the door and get better at that each quarter going forward.
Does the Wafer Services grow into Q4 or flat?
There's significant demand for our Wafer Services, and we're becoming more efficient with the way that we drive that Wafer Services revenue.
Got you. Thanks, Steve.
Sure.
There are no further questions at this time, and I would like to turn the call back over to Tom Sonderman for closing remarks.
Thank you. As I shared at the beginning of our call, SkyWater's TaaS model is perfectly aligned to accelerate the realization of disruptive ideas into the market at a time that demand for beyond Moore's Law technologies is growing to satisfy mega trends across industries. Additionally, a great opportunity is in front of our nation to act in pursuit of semiconductor sovereignty and to ensure the next generation of technologies are produced domestically. Our goal at SkyWater is to differentiate, disrupt, and then dominate the markets we choose to participate in. These factors drive our collective enthusiasm for the future of our company and U.S.-based high-tech manufacturing. Thank you for your interest in SkyWater.
Ladies and gentlemen, this concludes today's conference call. Thank you for your participation. You may now disconnect.