Racecar Project - Dyno Tuning
07-01-2006, 11:37 PM
#16
Nordschleife Master
Thread Starter
All dynos, regardless of brand or type or engine or chassis and will always have torque=HP at 5252 and Dynapack is no exception. If it doesn't, someone has fabricated the numbers, or are using a different measurement for HP such as PS or KW. HP = torque * RPM / 5252. The reason the graph does not appear that way is that the torque scale on the 'Y' axis on the left (81-270) is different than the hp scale on the 'Y' axis (24-330) on the right. If you look at the torque curve at 5252, you'll see it has 258ft/lb on the left scale, and if you look at the HP curve at 5252, you'll see it also has 258hp on the right scale. I could have certainly rescaled the graph so both 'Y' axis had the same scaling, then the curves would cross at 5252.
The Dynapack uses hydraulic pressure and valving inside to apply pressure against the spinning hubs to control axle speed, not eddy current. It allows for steady state tuning an all RPMs which I did to determine the proper AFR and torque for each cell in my roughed in table. In other words, I was able to hold the car at 8000rpm and full throttle and map it accordingly. This particular dyno will handle up to about 725rwhp and has been exceeded by only 1 3.8l turbo engine where it overtorqued the dyno and it couldn't hold it. It measures torque in .1ft/lb and is accurate enough to see a 1ft/lb change in torque resulting from a change to ignition timing.
If you look at the fuel graph above, you can see the reduced injector pulsewidth (less fuel) in the 3400-4750 range to maintain a 13.2:1 AFR across the entire 1500-8000 range at 100% throttle. I do not have provisions to measure EGT on this chassis dyno, and did not install the sensors to the MoTeC. As I've begun to think about this, the more I'm convinced it is an exhaust restriction issue. If I look at my timing table, the timing is lower in these sections (timing set to MTBT in all cells). If this were a case of engine ineffiecency, I would expect more timing, just like an engine at 50% load requires more timing. In my case, the ignition timing is lower in this range and the manifold pressure is at 100kPa, so I believe it is due to the exhaust system. The next test is to remove the mufflers and run with straight pipes. And YES, that is one ugly torque drop.
The engine is extreemly responsive with the small lightweight Tilton clutch. The entire clutch package - flywheel, pressure plate, twin disks and intermediate plates weighs less than a lightweight RS flywheel alone. And the mass is centered in a 7.25" space rather than the larger diameter of a stock setup. The clutch is much smoother than a puck setup found in the 964 Cup cars and it feels like the clutch in a Honda Accord. There are not stalling issues and the engine idles smoothly at 850rpm with very large camshafts, duration is almost 320 degrees. All-in-all, it is a good start.
Larry,
Yes, that is Rick's shop.
The Dynapack uses hydraulic pressure and valving inside to apply pressure against the spinning hubs to control axle speed, not eddy current. It allows for steady state tuning an all RPMs which I did to determine the proper AFR and torque for each cell in my roughed in table. In other words, I was able to hold the car at 8000rpm and full throttle and map it accordingly. This particular dyno will handle up to about 725rwhp and has been exceeded by only 1 3.8l turbo engine where it overtorqued the dyno and it couldn't hold it. It measures torque in .1ft/lb and is accurate enough to see a 1ft/lb change in torque resulting from a change to ignition timing.
If you look at the fuel graph above, you can see the reduced injector pulsewidth (less fuel) in the 3400-4750 range to maintain a 13.2:1 AFR across the entire 1500-8000 range at 100% throttle. I do not have provisions to measure EGT on this chassis dyno, and did not install the sensors to the MoTeC. As I've begun to think about this, the more I'm convinced it is an exhaust restriction issue. If I look at my timing table, the timing is lower in these sections (timing set to MTBT in all cells). If this were a case of engine ineffiecency, I would expect more timing, just like an engine at 50% load requires more timing. In my case, the ignition timing is lower in this range and the manifold pressure is at 100kPa, so I believe it is due to the exhaust system. The next test is to remove the mufflers and run with straight pipes. And YES, that is one ugly torque drop.
The engine is extreemly responsive with the small lightweight Tilton clutch. The entire clutch package - flywheel, pressure plate, twin disks and intermediate plates weighs less than a lightweight RS flywheel alone. And the mass is centered in a 7.25" space rather than the larger diameter of a stock setup. The clutch is much smoother than a puck setup found in the 964 Cup cars and it feels like the clutch in a Honda Accord. There are not stalling issues and the engine idles smoothly at 850rpm with very large camshafts, duration is almost 320 degrees. All-in-all, it is a good start.
Larry,
Yes, that is Rick's shop.
07-02-2006, 01:40 AM
#17
Three Wheelin'
Join Date: Nov 2002
Location: Winnipeg, MB Canada
Posts: 1,392
Likes: 0
Received 16 Likes
on
7 Posts
Wow.
As with all your posts Geoffrey I have to read them about 10 times to figure out 2% of what you are telling me.
Your car looks great and sounds fantastic. I can only imagine what it would be like without that exhaust on there.
Kirk
As with all your posts Geoffrey I have to read them about 10 times to figure out 2% of what you are telling me.
Your car looks great and sounds fantastic. I can only imagine what it would be like without that exhaust on there.
Kirk
07-02-2006, 07:41 PM
#18
Addict
Rennlist
Site Sponsor
Rennlist
Site Sponsor
My bet is that a set of 1 5/8" primary 3-1 headers with a reasonably conventional twin silencer set up will increase torque across the board and also eliminate the hole in the curve. My similar engine (3.92 with higher compression) runs practically the same intake system but uses stock 993 heat exchangers with 100cell cats & silencers and holds over 300lbft from 4000 to 7000rpms, so I shouldn't think that you will need to go far to fill in the hollow.
07-03-2006, 05:08 AM
#20
Addict
Rennlist
Site Sponsor
Rennlist
Site Sponsor
Provided that you do not have too many emission issues to worry about, IMHO fitting a well mapped Motec system is the best way to significantly improve the perormance and driveability of the 964 engine, the best (standard 3.6) engines making more than all the non-motec 3.8 road car conversions that we have seen.
Obviously I cannot speak for Geoffrey who will cover the costs in the USA, however in the UK we charge £3495 + VAT for the "basic" Motec conversion which includes:
Motec M48 ecu
9m M48 to 964 adaptor loom
9m Intake system
Throttle body conversion to TPS
MAP sensor
AT sensor
6 x Bosch injectors
Installation & dyno/road mapping
As mentioned on previous threads we usually see around 10% gain above a well remapped motronic 964RS and up to 17% from stock.
Obviously Geoffreys car is running the later M600 ecu which has a lot more functionality than the M48, but since my engine is still running the M48 and putting down over 350bhp at the tyres, which Motec ecu you choose is largely irrelevant.
Whilst on the subject of 964s and Motec, we also offer a more powerful package that we call the +1 which has the addition of 9m sport cams, ported heads & raised compression. Typically the +1 makes 20bhp more than the Motec on its own, but the point of mentioning it in this context is that we have a customer with a 964RS Motec+1 who occasionally does "quiet" track days in the UK, so he had made up a Supertrap tailpipe. Anyway, one day he complains that the car is not a lot faster than standard RS's and asks us to investigate, so we test it as it arrives with the Supertrap on. All I can say is that I never expected a tailpipe to lose 50bhp.......
Obviously I cannot speak for Geoffrey who will cover the costs in the USA, however in the UK we charge £3495 + VAT for the "basic" Motec conversion which includes:
Motec M48 ecu
9m M48 to 964 adaptor loom
9m Intake system
Throttle body conversion to TPS
MAP sensor
AT sensor
6 x Bosch injectors
Installation & dyno/road mapping
As mentioned on previous threads we usually see around 10% gain above a well remapped motronic 964RS and up to 17% from stock.
Obviously Geoffreys car is running the later M600 ecu which has a lot more functionality than the M48, but since my engine is still running the M48 and putting down over 350bhp at the tyres, which Motec ecu you choose is largely irrelevant.
Whilst on the subject of 964s and Motec, we also offer a more powerful package that we call the +1 which has the addition of 9m sport cams, ported heads & raised compression. Typically the +1 makes 20bhp more than the Motec on its own, but the point of mentioning it in this context is that we have a customer with a 964RS Motec+1 who occasionally does "quiet" track days in the UK, so he had made up a Supertrap tailpipe. Anyway, one day he complains that the car is not a lot faster than standard RS's and asks us to investigate, so we test it as it arrives with the Supertrap on. All I can say is that I never expected a tailpipe to lose 50bhp.......
07-03-2006, 09:42 AM
#21
Nordschleife Master
Thread Starter
Hi Colin,
As a guy who is more concerned with "area under the curve" moreso than peak power numbers I am a little embarrassed about the dyno graph since it is pretty bad. But, I feel confident it is not the engine design or components themselves, but rather the muffler. The headers are 1 5/8 stepped to 1 3/4 which I've run on other engines, including the 13:1 engine that produced 354rwhp. As you suggested in your example, exhaust systems, especially ones designed for "quiet tracks" are highly restrictive.
RE the MoTeC ECUs. The M4/M48 family of ECUs are current production ECUs from MoTeC, but they are the oldest. I still use a lot of M48s on 911s since the price/performance is excellent. The fundamentals of the M4/M48 and the M400,600,800 are similar and they both have the same number of sites in the tables. The newer ECUs just have more input/outputs and Windows software vs DOS software (which is simplier and I like better in some ways). My car has an M800 which is the highest ECU they sell.
As a guy who is more concerned with "area under the curve" moreso than peak power numbers I am a little embarrassed about the dyno graph since it is pretty bad. But, I feel confident it is not the engine design or components themselves, but rather the muffler. The headers are 1 5/8 stepped to 1 3/4 which I've run on other engines, including the 13:1 engine that produced 354rwhp. As you suggested in your example, exhaust systems, especially ones designed for "quiet tracks" are highly restrictive.
RE the MoTeC ECUs. The M4/M48 family of ECUs are current production ECUs from MoTeC, but they are the oldest. I still use a lot of M48s on 911s since the price/performance is excellent. The fundamentals of the M4/M48 and the M400,600,800 are similar and they both have the same number of sites in the tables. The newer ECUs just have more input/outputs and Windows software vs DOS software (which is simplier and I like better in some ways). My car has an M800 which is the highest ECU they sell.
07-05-2006, 07:18 PM
#22
Nordschleife Master
Thread Starter
Ok, I checked with the muffler manufacturer and they said that they have experienced the same type of dyno sheet on the larger engines like mine. They said the mufflers don't work below 5500. I also confirmed this with another shop that runs this setup for Lime Rock and they said my results were typical. So, back to the drawing board for the exhaust system. I want to quiet the car some from straight exhaust, but don't want to have such a problem like these. It won't get me passed the LRP noise limit, but I can install this muffler for those events. Looks like a winter project.
07-05-2006, 08:06 PM
#23
Addict
Rennlist Member
Rennlist Member
What is your target for the exhaust? FWIW, I run 1.625" OD headers (class rule) and Phase 9 mufflers (about 98 db) for normal use. For Laguna Seca, I have to swap to an older 911 style muffler to meet 92 db. They did MoTeC maps for each.
07-05-2006, 09:38 PM
#24
Addict
Rennlist Member
Rennlist Member
Geoffrey, What injectors are you using? I know that Colin says that stock 964 injectors are too small to achieve over 300hp.
It is certainly great being able to ride shotgun with you and your quest for more power. Thanks for taking the time for the complete write-ups.
It is certainly great being able to ride shotgun with you and your quest for more power. Thanks for taking the time for the complete write-ups.
07-05-2006, 10:06 PM
#25
Nordschleife Master
Thread Starter
Tom, For LRP I need to be at 89db max which this muffler does, but as you can see, there is a price. Ultimately, for the unrestricted tracks I run at, I'd like to be in the 100db range. I have some amount of hearing damage and don't want to run open exhaust. Even with earplugs, my ears ring aftwards.
I am running stock fuel pressure and 44lb/hr injectors. My peak injector duty cycle is 62%. Normally I try to get the max injector duty cycle in the 75% range, so these injectors are a little large, but not really an issue. If I had to do it again, I would choose a smaller injector and possibly run at 5bar fuel pressure. Below is a template from the MoTeC data logging that shows the dyno run. It shows engine RPM, injector pulse width and you can see it decrease where the torque falls off, injector duty cycle, and any acceleration enrichments. It also shows the difference between the effective and actual pulse width which is basically the injector dead time for a given battery voltage, in this case 13.9v. On the right it shows all of the channels I selected for logging out of the M800 ECU. I also checked the throttle linkage tonight as I was making a throttle pedal stop so as not to stress the cable and noticed that I was not getting quite full throttle out of the linkage. These runs were actually at about 95% throttle and I have some linkage issues to sort out.
I am running stock fuel pressure and 44lb/hr injectors. My peak injector duty cycle is 62%. Normally I try to get the max injector duty cycle in the 75% range, so these injectors are a little large, but not really an issue. If I had to do it again, I would choose a smaller injector and possibly run at 5bar fuel pressure. Below is a template from the MoTeC data logging that shows the dyno run. It shows engine RPM, injector pulse width and you can see it decrease where the torque falls off, injector duty cycle, and any acceleration enrichments. It also shows the difference between the effective and actual pulse width which is basically the injector dead time for a given battery voltage, in this case 13.9v. On the right it shows all of the channels I selected for logging out of the M800 ECU. I also checked the throttle linkage tonight as I was making a throttle pedal stop so as not to stress the cable and noticed that I was not getting quite full throttle out of the linkage. These runs were actually at about 95% throttle and I have some linkage issues to sort out.
