KPower Industries K24 swap

[RedReplicant]

Originally Posted by Irace86.2.0

If anyone wants to take a picture of the inside of their welds, I could tell if this was a penetration problem or a sugaring problem from not back-purging. The latter is a common shortcut.

Unfortunately, this is the only picture I took while I had it apart, and it was after ceramic coating.



Seems like no back purge to me

[captain awesome]

I’ve been running a deatschwerks fuel damper. Fairly simple piece but it works and was around $50.

[funkjaw]

Originally Posted by captain awesome

I’ve been running a deatschwerks fuel damper. Fairly simple piece but it works and was around $50.

Do you know which one I would need to ge? I see -8 or 3/8in...

[captain awesome]

Originally Posted by funkjaw

Do you know which one I would need to ge? I see -8 or 3/8in...

6-FPD-8AN was the part number. It should be -8.

[Irace86.2.0]

Originally Posted by RedReplicant

Unfortunately, this is the only picture I took while I had it apart, and it was after ceramic coating.



Seems like no back purge to me

Is that at a lap joint or just a butt joint? If it is a butt joint then the weld missed the joint, as the end of the pipe is still visible, but if it is a lap joint then it isn't bad, but you are right that it does look like there was no back purging. While there was penetration, even with the ceramic coating covering the welds, the welds look a little round and puffy like there was porosity. That is probably why we are seeing failures.

There are two ways to protect the backside of the welds. One is to back purge, which would require a separate tank or a dual flow setup, and it requires the appropriate plugs to run a hose to one side. This is more expensive and takes more time. The second way, which is what I have done, is you can buy Solar Flux or similar products and mix the powder with methanol. You paint the inside with the product, let it dry, and it will burn off with the welding process and displace the oxygen, so it can't react. It is cheap, fast, and it is pretty easy to clean up, so there is really no excuse not to protect the backside of the welds.

https://soulpp.com/2019/09/17/tech-t...-what-and-why/

[Ebush]

Has anyone else developed any gear grind or issues getting into gear since the swap? Initially I figured maybe I've just abused the Trans too much over the years. I bought the car new in 2017, and have only put 34k miles on it. During those initial 30k miles with the fa20 I never had a single issue with the trans, but in these last 4k miles I have an issue getting into 2nd gear quite often. Only at redline though, if I shift at 7k it seems to be fine. I'm starting to wonder if this Trans just doesn't like 8500rpm shifts. But it doesn't seem to have an issue in any other gear, so it could just be that I hurt that Trans at some point and it's starting to be more noticeable.

[Irace86.2.0]

I don’t have any issues. I don’t know if you need to adjust the clutch pedal, but outside of that, I can’t think of anything else it could be. I don’t know if the rpms matter.

[captain awesome]

Good thinking on the clutch pedal adjustment. I've had it off just a little on my previous FRS and it makes gear entry a little crunchy. No issues as of right now though.

[RedReplicant]

I had my car tuned today, assuming it make about the same power on the NASA classing dyno then it'll be about what I expected out of the stock motor on 91 -- 206whp/180ftlb.

My tuner seemed kinda disappointed, but my other friend made 225whp/191lbft out of his K24 Exige which seems to fall in line with what I made given drivetrain losses / temp difference / etc. Another couple points of reference are that my full bolt-on AP1 S2000 made 200whp/136lbft on this dyno and my EFR6258 Miata made 218whp/243lbft (massively turned down, car made north of 375whp on E). Classing dynos for those were 206whp for the S2000, 223whp for the Miata, and 222whp for the K24 Exige.

Anyway, a few notes:
1. Base TDC angle was 304* in the KPower map, Haltech lists 311*, my tuner got 300* with the timing gun. Who knows.
2. Knock sensor settings on the base map are basically unusable
3. We were only able to update to v3.05.1 firmware. Haltech support line confirmed that their stuff is broke af.
4. Tuner found that 90% was the highest power for throttle opening, this made a 6whp difference from the stock 100%. You can create a Drive By Wire Max Throttle Position map with your axes as dbw throttle position and RPM then set it to 90*
5. The base map target lambda table is rpm vs cable throttle position, so it is always 0. It needs to be changed to DBW throttle position.

Also, the car smokes a little bit (oil) at load so that sucks. I compression tested it a couple of months ago at 210psi on all 4 bone cold, so it is what it is. I'll move up my second motor build and just run it as-is.

edit: also, we pulled the intake tube and did a run, the car lost 10whp+ everywhere except the very peak 500 rpm where it gained a couple.

[captain awesome]

I'm interested in the DBW settings. With NSP I wasn't sure how to set it properly as it didn't look anything like the chart that's been shared on the FB threads. I tinkered with it a little one evening but only on the software as I didn't think what I was doing looked correct.

[RedReplicant]

Originally Posted by captain awesome

I'm interested in the DBW settings. With NSP I wasn't sure how to set it properly as it didn't look anything like the chart that's been shared on the FB threads. I tinkered with it a little one evening but only on the software as I didn't think what I was doing looked correct.

I have mine expanded to more cells than needed since I plan on restricting the throttle opening above 7000rpm on the dyno if needed for classing, but it looks like this:




There is a single cell by default, I think, so you could just set that one to 90. iirc, some people reported different numbers like 80-85%, so testing is helpful.

[RedReplicant]

Anyone looked at putting a K20Z3 / RBC or K20Z4 / RSP head on their motor? It seems like they share the same layout as the K24A2 head so our existing intake manifold and coolant neck would work.

[captain awesome]

Originally Posted by RedReplicant

Anyone looked at putting a K20Z3 / RBC or K20Z4 / RSP head on their motor? It seems like they share the same layout as the K24A2 head so our existing intake manifold and coolant neck would work.

This was something I dove heavily into. Almost bought an RBC and searched far and wide for an RSP. If you found an RBC ready to bolt in for a good deal, say $500 or so it may be worth it. I tried to find some dyno plots and saw that Kpower put together one in an E30. It mentioned only a 5 hp gain. Don't know if cams will make better use of the added CFM but it doesn't appear to be worth it unless you just had one kicking around.

[RedReplicant]

Originally Posted by captain awesome

This was something I dove heavily into. Almost bought an RBC and searched far and wide for an RSP. If you found an RBC ready to bolt in for a good deal, say $500 or so it may be worth it. I tried to find some dyno plots and saw that Kpower put together one in an E30. It mentioned only a 5 hp gain. Don't know if cams will make better use of the added CFM but it doesn't appear to be worth it unless you just had one kicking around.

Hmm, interesting, I'll see if I can dig up that KPower post.

It looks like new RSP bare heads are attainable, but it is pretty close to the price of a 4Piston CNC ported TSX head. 12100-RSP-000 is the part number.