I was talking to another guy who said he kept the r53 alternator and managed to keep the original pulley as well, I guess he flipped the alternator upside down to mount it and had to use a different belt.

I'm kind of surprised about the water pump though, considering it's a normal factory piece. Hopefully it'll keep up ok with a civic 1/2 radiator I plan on using to open up space in the front to put a vertical intercooler. I might try running a slightly cooler thermostat too and see if that helps. If it comes down to it, I'll just fit a fan temp switch directly to it and set the on/off temp manually like I used to do with my Honda's.

 

kinetic motor works in NC has also built a turbo r53 they might be able to shed some knowledge. while I have no info, I would love a turbo r53 so this is all helpful =)


soccerbummer is also doing a stupid highend conversion, but has been busy recently.


I would also be afraid of the r50 pump being enough flow, I doubt there is an easy option to change the impeller to get more flow. EWP might be the better option for warmer climates

 

Another thought.....the crank damper diameter is different on a r50...so the vanes in the waterpump are optimized to run and pump at a different speed.....haven dealt with the r50 much...but, here is a thought...
Run a pump too fast, and it create cavation, aka vacuum bubbles in a liquid......
Think the larger diameter of the r53 crank damper (making more liner belt speed) compared to the the r50 crank damper and waterpump is doing just this....the r53 crank damper is a smaller diameter.....so it might be more of an issue of changing the pulley to SLOW DOWN the pump so it can pump efficiently. Tried to find a quick source for the diameter of the r50 vs the stock r53, and didn't find one...but I seem too recall they are about the equivalent of a -8% (just a WAG) compared to the r53.

 

I don't think it would be much different than under driving the supercharger based water pump, and those are being spun 15-17% faster on average. From a quick visual it doesn't look like the internal physical design of the pump itself is much different, however the pulley on the R50 pump is MUCH larger and therefore slower overall. I don't doubt there is a different vane count in the R50 unit, but then again I don't think we'll have to worry about cavitation unless you plan on spinning the motor well over redline.

If you've never seen an R50 pump's pulley, a quick reference would be to look at the size of your a/c compressor pulley, because it's very close to that diameter. A few rpm's faster from an oversized r53 pulley shouldn't cause too much of an issue.

 

I actually did message KineticMW for more information and to see what they've run into in the past.

So far I've been told that the factory MAP sensor can see up to 20psi before it hits fuel cut, and that has been said by 3 people now so far. I'd say that's a pretty decent ceiling for the average joe just looking to swap over to something different. I still plan on running up to 18psi max though, rather be safe than sorry.

I've also heard from another individual (and I will be testing this theory once my build starts) that the upstream map sensor can be mounted into the intake tube and won't throw a CEL code. If it doesn't work, the code can be disabled through FA53 according to Kinetic so it's not a big issue.

 

Can anyone verify that the upstream MAP sensor is pretty only for use as a barometric sensor to provide a starting correction point for the main closed loop VE table? I have a really strong idea that this is what it's for, and if people end up trying to delete the sensor it will throw fuel trims off.

Think about it in this light too, every time the weather changes, the barometric pressure changes as well, so from a day to day basis your base VE fuel adjustment will change with the weather. I really do think that it's intended mostly as a baro sensor considering it's original location coming from the inlet side of the supercharger, and it could be a possible reason for some people's variances while using ByteTronik's FA53.

I'm also looking into whether or not the Mini ecu has a base VE table that can be modified for closed loop fuel control (and whether there is one for timing as well). Since a turbo conversion is going to change the VE of the motor a bit compared to the factory M45 setup (or even an upgraded charger), it's high on my priority list to find out if there is a base VE table that can be modified to bring STFT and LTFT into line as close as possible first (+/- 5% max) before pushing towards bigger boost tuning using the tables they commonly show to adjust for fuel/timing. As always, everything must be logged accordingly with actual wideband, STFT, LTFT, knock count, timing, etc. data to see the full picture.

My car has the EMS 5150 DME (not sure if it really makes a difference, I've only heard the newer ecu has more tables than previous pre-face lift models but can't confirm that), so hopefully someone can fill me in on what tables are available to edit.

 

any update with this build? I'm curious to see how it came out.

 

I actually finished my conversion over the winter and started it with the standalone a couple months ago. It's been running great so far and I'm finishing up fine tuning the 3-port boost control currently to hold a steady 21psi all the way to redline. I never put the build thread up over here because there didn't seem to be much interest in it to begin with, so I left it for a different forum.

For the size turbo that's on it, it actually pulls quite well, much quicker than the factory eaton did. That being said, I want to go with a larger turbo already than the little T03 50 trim that I used, probably at least a T03 60 trim at minimum or some type of hybrid. Next year I'll be building a bottom end for it, and getting the head redone as well, and looking at a much larger turbo overall. If you're interested though, I could copy and paste everything from the build on to a new thread over here so you can at least see what I ended up with.

 

What other forum?

 

I have it up over on TrackMini, it's more of a quiet group, but definitely advanced builds and fabrication over there.

 

Look up Ryephile...he did a successful Cooper S straight turbo conversion years ago...not sure if he's still here but you can find him at Michiganmini.org.

 

I've read most of his older stuff, he had a lot of issues back then that ultimately led to it blowing up on him. Most of the old photos and links are dead from back then unfortunately.

Thankfully, I pulled the trigger on a standalone when I built mine, and it's gone a long way to providing stable and steady power. Even throttle response is night and day with it compared to the factory ecu, not to mention having the peace of mind in knowing the ecu will do what you tell it to do.

 

If you're interested in seeing it during the initial first warm up that was running on a rough base map, I uploaded it to youtube a little while ago.

 

nice work! did you do most of the fab work yourself?

 

the R53 alternator had a different pulley option for the early gens. (the super early prototype mini's all used the same alternator and you can just buy the pulley.

It just simply bolts on, no shaving or grinding needed. (See my build thread.. somewhere.. I did it.. not sure if I took pictures or not. I also had to make a shim on one side of the alternator I think.. but everything else just bolted on and had good enough alignment)

As for anything else.. just go for it.

I would suggest (and I am going to do at some point) making the radiator shorter, deeper, and a double pass, and putting a FMIC up top, and an oil cooler down low in front of your AC condensor.

I have worked out a way to make it all fit.. I think.. and not shrink the cooling capacity of the radiator setup when compared to stock.. (taking the oil out and putting it in it's own loop with a thermostat to reduce some of the thermal load on the glycol cooling loop and radiator)

 

And it wasn't really a baro sensor, as that should have no effect for a speed density. Air mass going into the engine is still a function of pressure in the intake manifold and temperature of that air. (for a given piping system pressure differential is what determines gas velocity, and air temperature determines the mass flow rate of that air stream at it's given volumetric flow rate)

what I think the upstream sensor was used for, since it was still post TB, and looking at the pressure curve and the map values for both sensors when I did some bytetronic work was to look at the delta P across the supercharger and make sure it was within range of what the car should be seeing.

If the fuel being added had to be trimmed out too far based on the pre and post map sensors it would trigger a fault for the BPV or the SC itself.

Why? Because a supercharger is a positive displacement pump, which means, more or less, that at a given pressure ratio across the supercharger, it should have a fairly set volumetric flow rate. Combine this flow rate with a intake temperature in the manifold and you should be able to calculate a very close ballpark figure on how much fuel you need. It sort of served as an extra redundant version of speed density.

If the car ever went too far out of parameter it could swap to a plain speed density setup. (just look at the TMAP in the IM and calculate fuel and adjust)

so in a way it is a baro correction.. or it functions like one, but it is not adjusting for barometric pressure, it is adusting the pump curve of the supercharger based on inlet pressure (post throttle). Pre throttle would be strictly a baro compensation table)

 

So far my 3 core honda radiator has worked very well (surprisingly). My average cruising water temps sit around 74c, and at full boost (24psi at 9.5 degrees peak torque timing around 3400rpm, timing climbs back up to around 19 degrees at a redline of 7100rpm) I only see a max of 91c on multiple back to back hard pulls. If it was a dedicated track car then I would definitely go for a longer radiator, but as of right now I can't go with a thicker core as it physically won't fit (currently 3.5", any thicker will hit the intake manifold).

I gave up on the whole Bytetronic idea early on. I wanted repeatable performance no matter what, and from everyone I talked to the factory ecu was not that option. Changing the air flow the upstream map sensor was normally used to seeing (I think baro was the wrong term to use, I figured it was to calculate flow into the supercharger inlet in comparison to manifold pressure and calculate fuel and timing accordingly) tended to cause a lot of issues, mostly timing related during cruise speeds and rpm (people reported only being able to reach 16 degrees advance during cruise, in reality it should be around double that). The reality is, a standalone is the only real current option for a platform that is over a decade old now.

I ended up having to replace the turbo after a few weeks just due to poor balancing of the one I had on hand, and went to a larger "super 60" type T03 with an upgraded compressor housing and blade design. It's older turbo tech, but so far my budget was $3,000 and even with the new turbo and standalone I managed to stay within that goal. It's not a pretty setup, but it's definitely fun for the money and still edging close to that 300hp range.

 

I have a buddy that is moving toward a turbo with a local shop here in NC. Yesterday we were at a C&C and he was talking the K4 that the Mazda 3 uses. He wants to get some summer driving in before it goes to the shop.

 

I'm not that far from you either heh.

The twincharged 'project stupid' car was at the NC halfmile shootout under my buddies shop flag earlier this month. (Performance MD out of Charlotte / Rock Hill)
.. so was an Underground Gallardo... Both cars got some looks. bahahaa.