I would note a couple of other things just really quickly before I turn it over to Neill. We have talked about a dozen or so parts that are qualified out of Mohawk Valley, including our biggest parts, and most complicated parts and so forth. The other thing to realize is all of these MOSFETs that I just talked about qualifying, they qualified first-pass. And that is not normal. There’s — you normally find something, some problem or qualify with an asterisk saying you’re going to put some kind of something in place to kind of make sure that the quality that we only ship quality products to customers, we qualified first-pass on all those parts. And I think that gives us a huge boost of confidence in the underlying capability of this factory.
Neill Reynolds: Yes. So Harsh, maybe it’s important for me to just break down the gross margin a little bit and some of the markets that we’ve had out there and kind of maybe just give you a little bit of kind of my view in terms of what that means from an underutilization perspective. So we had a good quarter from a gross margin perspective above the midpoint of our guidance range. We’ve also seen strong underlying gross margin improvement, excluding the underutilization over the last couple of quarters. In fact, if you go back to the June quarter until the guidance we just gave here in Q3, we’ve seen approximately 400 basis points of improvement of margin in just three quarters on an underlying basis. In addition to that, I think we’ve seen the cost on the 200-millimeter substrate, the yields on those as well as the initial costs in Mohawk Valley on a unit basis be very much in line with what we had anticipated.
So as we’ve always said, the more revenue we pushed through Mohawk Valley the better we’ll see — the faster we’ll see that kind of some of that gross margin execution. Now as you said previously there on the underutilization, that will be a little bit of a drag for us. We talked about $36 million last quarter at Mohawk Valley would be flattish and get up to a peak of about $38 million as you get into Q4. So what that will mean is, we kind of put a marker out there of about a 20% gross margin or approaching that in kind of the Q4 time frame. You can think about that in the kind of mid to high teens, probably exiting the year and then back maybe pushing up 20% or so out of quarter. So as we look out in time. So what that means is, yes, we will start to see the underutilization kind of peak out there in the Q4 period start to come down a bit.
But just a reminder, those start-up costs you mentioned in Siler City, those are currently on the OpEx line, about $10 million last quarter. That will probably grow to about $15 million as we exit the year. As Siler City goes into production probably in the second half of fiscal ’25, that kind of first half of calendar ’25, we’ll see those transition from underutilization up to the gross margin line. So a few moving pieces there to think about, as I mentioned in the prepared remarks, but the key here is underlying we are seeing very strong performance and anticipate seeing that strong performance continue as we drive more volume through the scale facilities that we’re building.
Operator: Thank you. The next question is from the line of Jack Egan with Charter Equity Research. You may proceed.
Jack Egan: Hey, guys. Thanks for taking the question. I just had one. I was curious, why didn’t either of the recent LTAs signed with Infineon in Rome include 200-millimeter, because we’ve seen some other companies sign some wafer deals that have included 200-millimeter even if it’s just a general long-term aspiration to eventually move to 200. So I mean, wouldn’t the JP at least eventually give you some capacity that you could turn around and sell to device or materials customers by, I guess, 2026?
Gregg Lowe: Thanks a lot, Jack, for the question. And I would start off by saying all of our long-term supply agreement customers are asking about 200-millimeter. It’s obviously front and center in their mind and something that they’re very interested in engaging with us on. We’ve engaged so far. We’ve announced the engagement so far of one, which is the deal we did with Renesas last year. And just to recall, that begins shipping 200-millimeter in 2027. In between now and then, we’ve got all of our focus on getting the JP up and running to feed the Mohawk Valley Fab. Getting to 20% utilization in the June quarter, got really good line of sight, but we’re also going to have to ramp the JP to get the other 80% utilization. Now the one thing that I would tell you is, we have a high degree of confidence in our ability to do that, and I’ll tell you why.
In March of this past year, we turned on the Building 10 operation. We’ve got certificates of occupancy, and began producing wafers and boules and wafers out of that facility. And less than a year later, we’re feeling very confident of taking what was a basketball court, a squash court, a bunch of offices and converting it into a 200-millimeter silicon carbide crystal growth operation and being able to feed 25% of the world’s largest 200-millimeter wafer fab. Well, all of that experience is going to be applied to the JP, which we purposely kept less than an hour’s drive from our campus. And the reason we did that is we wanted the same people who ramped Building 10 to ramp the JP. And so now you’re talking about a purpose-built facility. We’re not trying to work around what used to be an office and all the — you can imagine all the challenges we have had converting that space to a crystal growth operation.
