Jim,
Nice bracket... maybe that's what I'd need as well.

I re-located mine to where the engine thermostat is. But I still have to skew the temp curve to make it go to ~160F operating range though the temp is more stable in that location.

Which cylinder did you mount the sensor to?

BTW - great progress!!! Keep us posted.

Hobbieboy, I was thinking about your intallation in the T-stat housing, but I found the warm-up enrichment is only required for the first 2-3 minutes of running (60F ambient at start up) so I looked for a fast acting location for the temp sender.

The temp probe clamps onto the rear most edge of cyl #1 cooling fin just as pictured. Takes less than 5 min to remove the air duct if I had to get at the temp sender.

I mapped for 23% enrichment at 60F and tapered to zero enrichment at 122F. Any enrichment beyond this temp was bogging the engine down due to being too rich.

The operating temps I see at the location where sensor is on the fin have not gone under 152F while driving, and the hottest Iv'e seen is 205 while driving up hill at a continious 4 psi boost.

Seems as though the air cooled heads warm up pretty quick and with the injectors so close to the head casting, enrichment is only required for a short time. I was thinking about a water cooled engine where there is far more mass in the head casting and with the water jacket equalizing the temp throughout the casting, warm up would take a much longer time, perhaps 10+ minutes of running or more until you would reach a head temp where the mixture would not condensate out on the port wall.

Jim

What TPS sensor did you use? Is it a Porsche TPS or from another car? I'm trying to find a spare for my TWM throttle bodies and the one you used looks very similiar to mine. Any part number by chance? Thanks

The throttle position sensor came with the system from SDS. They give you the choice of either rotation direction, but I don't know of any part number or source. The installation manual describes how to interface with several OEM TPSs too.

Jim

umm... I really need to try your location then where I mount mine now, I still see a 20degF difference depending on whether I'm moving or stop @ traffic light.

I do agree that once my oil temp needle (dashboard gauge) rises above the bottom, warm up enrichment can be greatly decreased or it would be way too rich.

Jim,
How's the intercooling coming along?

Any plans for a dyno run?

Great project!

Doug

I just started looking at intercooler construction a couple days ago. I picked up some mandrel bent elbows and straight tubing, and prepped one of the cores. The basic conept of my intercooler design will duplicate the factory 1974 turbo RSR concept only in a more subdued style for the street. I'm in th emiddle of helping a friend with a 3.6 transplant so the intercooler will be on hold for a little longer unless I get some spare time, but I do plan to hook up the wideband O2 today, as well as take some pressure differential measurements above and below the 3.2 throttlebody to determine if it's size is a restriction.

Jim

Update

I completed mapping the fuel portion of the SDS EFI computer and ran a track day last weekend. The whole system is working great, with surpisingly little need to alter or modify anything. I thought there would be more work ahead.

As mentioned previously I started mapping with a narrow band o2 sensor then converted to a wideband for comparison. The readings from the wideband showed AFRs in the low 13s at light to medium load, but only up to about 3500 rpm then leaning as rpms increased. A data log run at WOT showed mid 12s up to about 4000 but then leaning to mid 14s as the engine approched 6000 rpm. After seeing the accuracy of the wide band I must admit it is the way to go. More initial cost, but certainly worthwile for a performance application.

I increased the fuel values in the upper ranges of the SDS rpm chart which will fatten the mixture under light load – mid to high rpms. After another data log run at WOT I calculated that I needed 4% more fuel at full boost – mid through high rpms. I did one more data log runs with the enrichment knob turned +4% and the injection sustained mid 12s to high 11s right through to redline. I adjusted the fuel values accordingly but only adding 3% instead of 4% to the MAP chart and repeated a data log. The result was mid to high 12s at 10 psi from 2500 through 6500.

I took the car to the track this last weekend and and continued to make a few minor adjustments and now it runs consistant low to mid 12s throughout. I'm impressed, the car is hauling A$$. I was seeing near my old straightaway speeds with only 10 psi which previously would require 1 bar. Hmmm, can you say, manifolds!

One issue has developed, but it's a good one. The 42lb injectors which I felt would be sufficient to support my conservative approach when I estimated I'd only be making near 400HP, are too small. The SDS has a gauge mode which displayed just over 90% duty cycle at 10 psi and 6500 rpm. No margin at all since the manual suggests no more than 85% duty cycle. Compounding this limitation is the intercooler which is yet to be added, and I thought one bar would still be in the cards. So, 72lb injectors come next.

Question
I'd be curious to hear if anyone running 72lb or larger injectors has developed idle smoothness issues.

Here are some basic calculations:
(42lb injectors x 6 cyl divided by 0.6lb/HP/Hr = 420HP) at 90% duty cycle= 378 HP

Same calculation using 0.55lb/HP/Hr = 412 HP (stoich)

I estimate the engine is between 378 and 412 Hp with 10 psi boost, and no intercooler. Yahoo!



Jim

Jim, have you thought about using 550cc/min (52lb I think) injectors? I used the calculator from here - http://www.rceng.com/technical.htm#WORKSHEET

It looks like 52lb ones will be sufficient. I'm using those and I'm only seeing max 65% duty cycle so far to maintain 12.x:1 AFR though mine is far from done (having MAJOR ignition problem).

And buddy of mine who uses the SDDS system (his is the 934 listed on the site) also uses same injectors for years without problem so far. His makes ~500hp.

I'am using Siemens 55 lb injectors with a rising rate Malpassi fuel regulator with no problems. Idle is reasonably smoth as well.

Great news! As I said before, narrow-band lambda has no use for tuning EFI. It's good for running stoich on cat-equipped car, that's it.

About injectors: there is no danger for engine even if you exceed 80% duty-ratio. Only thing that is important is AFR's. If your AFR's are in check then engine won't ping. That being said, running 100% duty-ratio will let current flow trough injectors all the time, which might overheat or fail them. So hell yeah, step up to bigger injectors. With intercooler and more boost, you will go trough the roof if you already run 90% @ 10psi and no intercooler. You will need real peak & hold injectors + driver though, or idle might suffer.

That being said, how's your ignition map? AFR's are one side of the story...ignition is another.

Good map will allow you to extract more power from the engine with less boost (= more efficiency and less compontent stress)

Which WBO2 did you use? LM1? Do you have access to AuxBox? If yes, do a couple of RPM vs. accelleration log runs while fiddling with ignition advance and listening for ping. Try to get optimal accelleration (thus torque) trough the revband, then back a two-three degrees for safety. It will give you max power. It's best for you to do this after intercooler is installed as it will give you extra margins.

This is when an hour on the dyno would really do some good. It could be disastrous to hope to hear the detonation. A good tuner and load dyno can dial in the timing without endangering your engine. It's pretty easy to advance each cell until the power increase begins to plateau and before detonation is a risk.

Yeah, dyno is nice if you have the funds. When mapping ignition, it's primarily the torque you look at, not power. Power is just a function of torque and revs. You look for max BMEP after which torque starts to drop down.

There are many variables here and you might run into ping before torque reaches it's max theoretical value. It will depend on air intake temp, twin or single plug, AFR's, combustion geometry and heaps of other factors. Some cars are knock-limited, some will allow you to go past max BMEP w/o knocking (thus allowing you to see that you actually passed the plateau).

By doing road runs with acccellerometer, you are actually measuring torque, thus doing exactly the same that you do on the dyno but with car in it's natural habitat (airflow trough intecooler, air-temps etc.) Arguably, sometimes it's easier to hear pinging from sitting inside the car than standing in dyno-room with ear-plugs.

Also, doing a road-runs is free.

Well here in the states, dyno time is $125-150 hr, not much more than the Porsche service rate. When you spend as much time, effort and dollars on an engine build/conversion etc, why stop there. The dyno time would be significantly more rewarding than spending it on strut braces and the like.
I should add, that in this case, Jim has a bit more understanding of the efi system than most as his installation well illustrates:)

I agree with Goran about using a dyno... lack of airflow to the turbo/intercooler will reduce the usefulness of tuning on dyno. My mechanic, who has an engine dyno says the same thing even for NA cars - you can get a good base map but nothing like real driving condition.

Having said that, it is virtually impossible to find a stretch of road to tune the top end without being put to jail here ;)

With many thanks to Jim2, my 3.6 swap is completed! Could not have done it without him! I am sure the turbo junkies will be glad to hear that, so he can return to his efi conversion, and what is one of the best threads I have seen in a while.

Hi all, here are the "nuts and bolts" regarding my injector sizing issue. Before tracking the car I installed a thermocouple in the charge pipe which indicated 190 F at full boost. After discussing this with SDS they calculated that about 15% more fuel flow will be required if my new intercooler holds my inlet charge temp to about 100 F. Regarding boost, I intended to run up to 1 bar which they calculated would require an additional 20% fuel. Together these indicate that 35% more fuel will be required: 42Lb (current injectors) +35%=56.7 Lb

There are several other factors which mess with the calcs, I/C effectiveness at 15 psi, extra RPM etc.

If my current 42Lb injectors are seeing 90%+ duty cycle then I should build in a little margin to reduce the duty cycle to 85% per SDS or 80% per RC Engineering. I'm leaning towards the next increment above 56.7Lb which seems to be 72Lb.

Per all of your posts above I'm going to buy some dyno time to dial in the timing but not until after the intercooler is done.

As mentioned very early on, I run a knock sensor which I continually monitor, likwise EGT. I suspect I probably have my timing close to optimum, but the dyno runs will tell where I'm at.

Here is an intersting observation. SDS was wondering if the 3.2 throttlebody was too small for the application. I put a pressure tap above and below the TB. Results show 2.5" Hg (1.25 psi) differential at WOT 10 psi boost. I'm wondering what it will be like at 15 psi?

Well, so much for my "conservative approach". I've got the bug!

Jim

Congrats Jim on THE best thread on these boards at the moment! Ive only just found the thread (no spare time to myself any more-baby number 2 has taken care of that!). Im looking to use SDS for my 930 engine so seeing/reading this thread is great! Your quality of work is 100% It actually irritates me a bit because I know for a fact that my install will not be as nice as yours! keep up the good work, Im glued to this thread (along with lots of others)

Where did you get the absolute manafold pressure gauge that goes to 70 inches? I've been looking for one of those.

That's a manifold pressure gauge for a twin engine aircraft.