When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Took it to the dyno, headline is it didn't make anymore power. Made 345bhp / 309whp compared to 352/3bhp / 312whp. Made the same peak torque at 265lbft. What is interesting is it made this peak power at 7200rpm rather than 7400rpm, and the power curve actually rolled over. We revved it up to 7800rpm but it didn't make anymore juice. We tried adding a bit of timing but that had no effect, then reducing timing decreased power, so that feels pretty optimised now.
Interestingly the car was more responsive to fuelling changes this time around, it was running a bit rich when it first arrived and we pulled fuelling back to 11.5 to 12afr and that got us back to the 345bhp figure from 325bhp it came in with. Previously the car didn't mind running low 11s high 10s. On the road this translated as the car getting its zing back. I gave it a good pull up a hill on the way back and it felt awesome again like it did when I first put the new cylinder head on early this year. I think with the faff with the water injection, rebuilding with a few changes etc it had just fallen out of tune a bit.
The changes this dyno session vs last were:
> +1 intake valves - haven't made a difference, maybe they require some more cylinder head to unlock them.
> New manifold and inlet for throttle body - hasn't made a difference over last time.
> New chargecooler. Has lowered PR everything else the same, increase in air temp over a dyno pull has reduced from about 10degC to about 2-3degC, hasn't made any change to power, atleast on a cold winters day.
We moved the intake cam around and actually found this time we made slightly more power at around 108atdc (max lift), where before it preferred more advance (102atdc max lift). Retarding the cam increases boost pressure. But as it made more power at 108 I guess 102 actually just had too much overlap, just blowing charge straight out the exhaust, lowering boost but wasting air and fuel. I sort of wish I hadn't added +1 intakes because I think my original standard size but aftermarket intake valve head from last year was actually a bit of a sweet spot. Maybe the +1s need a bit more of an aggressive deshroud to get them working for this application. Anyway I feel like the development has run its course now.
I also made a larger pulley for the supercharger, to go back to the 2.2 ratio I ran on the very first dyno session in 2020. When I added my 2.5 ratio pulleys they didn't add any power at that time, and if you've read this thread from the beginning you'll know I've never been 100% on why. So I was interested to see if the car would just go and make 350bhp at 2.2 ratio this time around.
It didn't. It made 327bhp / 288whp / 240lbft, at about 0.7bar of boost. I'm pretty happy with this. It is a lot of power for the boost level, without overspinning the supercharger. It will only be outside its continuous rpm rating above 6800rpm, and well under the intermittent rpm for the sc. This will really help the supercharger on a hot summer trackday, without water injection through it. Also, as the map used to pull 1-2deg of timing by the end of a straight due to AITs, if the larger pulley avoids this then there isn't really any trade off on power anyway. I think this is why the 2.5 ratio pulleys didn't make more power originally. It was a really hot day, the ECU pulled a couple of degrees, and we didn't see the power. Wish I had taken logs from those sessions but there we go. But mystery solved I think.
I had a track day today in the car, but to be honest nothing to report. It was really foggy and they couldn't open the track up properly. Got one run of 3 laps. It was very cold. With the new chargecooler aits haven't shifted at these low ambient temps. On the move, the car runs air temps of 25-30degC. This is basically the case from ambient temps of 15degC and below. I guess it is because of the way the setup is bolted to the cylinder head that also runs at basically the same temp independent of ambient conditions. Below is a comparison of a run on the same track on another cold winter trackday from 2021. I was sat in a queue to start with hence the higher starting temp but you can see overall the temps recover to the same sort of values. Log from today is bold, faded is from 2021
So basically, larger chargecooler does decrease temp increase over a pull, but overall at low ambient temperatures the overall average running temperature hasn't been affected. Will be interesting to see the data when the weather gets warm again. Will feedback on that next spring / summer.
Anyway, I am going to stop messing with this setup for a bit. I feel like it has reached a zenith and does exactly what I want it to do, and that I have explored the various paths I wanted to. If you follow my facebook or instagram (Coldside Engineering) you will have seen I have started to build a kit around the TVS (HTV) 900 supercharger which is to sell, which is keeping me pretty busy. Setup will be like my car, coldside, water to air chargecooled, performance should be equivalent to the setup I have developed in this thread (will be dynoing), but engineered not to be such a pain to install, keeping the alternator where it is, the ability to run ac and ps, not having to move the clutch master etc. It is a 2 core chargecooler, to keep things a bit more compact, with a nice cnc'd manifold that I am pretty jealous of and may have to get one machined for my car. So get in touch if you are interested in one of those. Saweett!
That's a short nose blower, non-MX5? Or is it repositioned in relation to the pulley plane? Oh wait, do I see Harrop there?!
Great work, always been a fan of these since a local had one on his NA, and the competition was the JR kit.
Thanks Gee Emm. Yes it is a Harrop manufactured HTV/TVS900, it actually has a fair snout on it to be fair. It is positioned to run on an additional pulley mounted in front of the crank pulley, rather than my car which has its supercharger inline with the old ps / ac belt run. I did this to allow use of ac and or ps. Also allows you to slot a tensioner plate in between the alternator belt and the supercharger belt - means the alternator can stay in position.
Thanks for that info, I knew they were doing installations for the big bent 8s here, so seeing the name was not a total surprise. Will be following with interest, though I think my caravan has departed as regards my FI commitment to the current build.
Thanks for that info, I knew they were doing installations for the big bent 8s here, so seeing the name was not a total surprise. Will be following with interest, though I think my caravan has departed as regards my FI commitment to the current build.
But never say never ...
Yes most of their kits are for big v8s etc. Well, get in touch if plans do change.
Originally Posted by Labora
Very interested to see how the TVS900 setup turns out.
New guy here and I hope you have enough pity on me to help me with my issue.
I have an older Kenne Bell supercharger and I am trying to take it apart. I've taken the bolts out, but I cannot separate the housings. I suspect you had the same problem and I am assuming it's just some sealer/Loctite. There does not appear to be a place I can get under to pry it loose. What was your trick to get them apart? Heat?
Replied to your DM Butcher, but I've never had my supercharger apart personally. I do know there are some internal components in the rotors that need to be undone from the gearbox end before splitting. But I've never been inclined to roll the dice myself as my rebuilder Pete is so reasonable. I know enough to know I don't want to mess with it
I've recently got the TVS900 kit on my car for testing. It is a bit of an animal. I am spinning it slightly faster than my Autorotor, as it has a higher max rpm, but the boost curve is certainly fatter from lower rpm than my 2087. So much so that it slipped the clutch in the midrange, and I have had to put an angrier clutch in it to hold the extra power. It has also over maxed my injectors, I am at about 97% duty by 7000rpm. So I will wind up the fuel pressure a little to get me by. It is a little quieter than my Autorotor in the cabin, but still makes a great noise. Still very much supercharged. Apart from the clutch and the injectors, only snag with it is the idle valve. The throttle body I bought has two tiny m5 ports for qn idle valve. I tried to use them but no particular surprise they were completely inadequate. So will have to come up with something else but no particular drama. On the dyno next week unless it snows.
I did reply back. If I ever do find an answer, I certainly will be back to post what I found. There are rebuild 'kits' available and I suspect there is only one way to do it right. I am just wanting to take the lobes out of the housing to clean the housing out a bit.
Got the fuelling dialled in and it made 365bhp, 326whp, 281lbft of torque. Very happy with this. Was a bit lucky because the barometric pressure was quite low, which bought me the duty cycle on the injectors to just about be ok, running them right up to and slightly over 100% duty. It also bought boost down a little to roughly 1bar in the top end vs the 16+psi I was seeing in the high pressure weather when I first got it on the car.
I've spliced in the torque curve from my Autorotor into this graph, sadly power was on a different scale
It's a decent wedge over the Autorotor. I am spinning the TVS faster - using a 2.57 ratio rather than 2.5, but the TVS is still under rpm limits by a decent margin whereas I am overspinning the Autorotor.
TVS is making 6% more torque about 3% more pulley ratio. It is a very impressive unit, I am a bit gutted to be selling it to be honest. You can really feel the extra torque on the road. Took the long way home and it was hilarious.
The chargecooler is working well. It uses 2x 45mm Laminova cores. My old 2 core used 2x 40mm cores. I was gaining about 9-10deg in AIT on a pull starting from 2500rpm up to 7500, the old 2 core is essentially the same, but it was dealing with a lower PR, less boost, less power. We did some more extended runs with several pull linked together to try to simulate some track driving, and it did great. It is interesting, once it gets up to temp, air temp actually starts reducing on a pull and it runs hottest on the overrun! It wasn't a massively long run, but only gained about 2degC over starting temps and it had stabilised. So happy with that. It was a cold day but, these setups tend to run at about 30degC base temp whatever the weather. The intake temp sensor is so close to the engine I think it picks up the heating effect on the air from the cylinder head. Then that base temp stays pretty much constant up to about 15-20degC atmospheric, then you start seeing the atmospheric temperature adding to that base temperature by about 20degC. I was wondering if I should have done it with a 3 core, but the extra width that requires made it a far more difficult install, so I am very happy to see the performance of the 2 core being so good.
12-02-2023, 05:03 PM
2
