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.
A bit of a techie post but it expands a bit more on the discussion about torque. We went back to the dyno yesterday for 4 hours and another 40 pulls. This makes the total number of over 120 pulls as part of the Motec tuning process here. I figured we were pretty much done after last weeks dyno session but Chris scrutinized the logs in details and was not quite happy that everything was perfected. He studied the XR1000 compressor map which convinced him to make a few tweaks. He ran the calculations and plotted the engine output onto the compressor map as can be seen by the straight line with rpm reference points.
He also wanted to finalize the knock sensor tuning and tweak transition fueling when quickly coming on and off the throttle which was not quite to his satisfaction. After about an hour it took him to complete this, he then wanted to set up maps with flat torque curves for maximum drivability. He also wanted to establish a practical and safe tuning limit for 91 octane pump gas. With 91 octane in the tank, Chris went to work doing pulls and scrutinizing timing and any signs of knock which was being monitored live. Amazingly with 91 octane, we managed 981whp / 727 wtq at 1.35bar which is absolutely mind blowing considering pump fuel. We were 20hp away from the 4 digit barrier. Ahh, where is that better Y-pipe? Adding any more boost beyond 1.35bar past 7000 would have the ECU start pulling timing so 981whp was the practical / safe 91 octane limit.
Since we normally run 95 octane at the track, we decided to add 5 gallons of 100 octane to the 5 gallons of 91 that remained in the tank giving us the 95 mix. Chris then added another 0.1 bar boost at 7000 and beyond. Another quick pull and voila, here is 1015whp / 745 wtq at 1.45bar. Mission done. The dyno pull video below really shows how linear and effortless this pull appears.
Finally, we also wanted to establish where we wanted to set the practical torque limit for track use for maximum drivability and engine longevity. When I ran the XR980s, I found the car very easy to drive when torque was limited to 600 ft.lbs as one could aggressively come in with power on turn exit without compromising traction. This level of torque is also very easy on the engine internals. As such, Chris wanted to make a file with a flat 600 ft.lb torque curve which will make for a super linear power delivery. I don't believe a curve can get better than this. This map smoothly ramps up boost with 0.7bar at 4000, 0.8bar at 5000, and just barely kisses 1.1 bar at 7500. The engine pulls strongly to 7800.
Sharp eyes will notice that on the 981 and 1015 graphs, the hp curve peaks and starts to taper off at 7200 and 7100 respectively while on the 858whp dyno the hp curve does not begin to fall off till 7600. What this tells is the fact that the engine begins to fall out of optimum efficiency as the HP approaches 1000. Looking at the logs clearly confirms this and you can also see this at the 7.7K reference dot onto XR1000 compressor map. The Turbine Drive Pressure Ratio (MAP/EMAP) begins to fall below 1.0 earlier in the rpm band as hp increases. At 858whp, the MAP/EMAP does not drop below 1.0 until 7200 while at 1015whp it drops below 1.0 at 6700. As such, in its current form, this engine is essentially at optimal efficiency all the way to redline at up to 875whp or so. By comparison, with the previous XR980s and OEM headers this 1.0 crossover point was at 4500 and down to 0.8 by redline.
One last thing we did was test the Y-pipe by adding a pressure sensor to it at the narrow right leg and could see at close to 0.1 bar pressure drop there. This reading would have been more precise and greater if we had a sensor at the intercooler outlet pre-Y-pipe but we did not have time to do that. With that said, it appears that a larger Y-pipe will free up another 30-40whp or conversely allow for the same output with less boost and less heat, again raising the engine efficiency. Chris and Johnny have already drawn up a Y-pipe on CAD. It will consist of 3.0" legs and a 4.0" collector pre-throttle body. The center section will be CNC machined and the 3.0" legs welded on. This will be a perfect no compromise bolt in swap in place of the OEM Y-pipe.
Finally, we tested the restriction the OEM airbox. We installed a large open tube with a bell mouth as pictured below in place of the airbox for a quick pull to see the restriction. Once the was done, data showed that the while the 100mb airbox restriction was now gone, the 100mb restriction had now shifted further down stream to the intake pipes which before showed a negligible 35mb drop when used in concert with the OEM airbox. To see an overall improvement there, one would have to replace both the airbox and intake pipes together for something better. In a nutshell, this showed is that the airbox in and of itself is pretty darn good when used in concert with the large intake pipes.
Looks like we are going to do one more full day on the dyno for final tweaking and verification of everything and call the engine tuning exercise complete. It's been quite the journey so far. Pretty educational and eye opening. Cheers...
Last edited by powdrhound; 10-27-2022 at 03:47 AM.
Absofreakinmazing! Thank you for all the research testing and sharing with the community. Knowing what y'all know now, how much of this can you reasonably see trickling down into how folks with closer to stock builds/power changing their power upgrade path in terms of improving engine efficiency before adding a ton of boost?
To further clarify, how might one choose their upgrades to get a similar flat torque curve like yours but at a 500-600 wheel hp 400-500 wheel tq?
We JUST got filters and venturis back in stock (they were backordered for more than a year due to manufacturing issues) for our intakes. I am still waiting to receive Bosch MAF housings in to complete the bolt-on 996TT V-Flow, but if anyone is running MAFless or wants to run our Hitachi software, I do have Hitachi housings available and can come up with a plug for the sensor port to allow the rest of the intake to be used. The V-Flow and our inlet ducts work well up to at least 700whp based on recent testing; in the past we have used them up to 850whp though I believe the filter and OEM upper duct are out of their league at that point. Performance vs. compromises of use and installation, I think they are a solid solution for that power level compared to the alternatives.
Sorry to anyone who has had a poor experience with us for the last couple of years. Between the pandemic, a move, and some personal issues on my part there have been a lot of bad bounces especially related to having parts available. I had to lay off a couple of full time staff and take a giant load onto myself during this period due to the losses of revenue from the backorders and the investment required to retool and restock our product line--frankly, I've done a far from perfect job of customer service throughout that situation. But my team and I have worked really hard to get most of our really solid, proven parts like our 996TT kits back into production and the higher volume items reliably stocked on the shelf. While that is still a work in progress, big parts of that plan have been completed and we're on to things like bringing on more support, revamping our website, and designing/testing/producing some new products and kits even for the 996TT.
Sam
Sam,
As it seems to be the way with all engineering/manufacturing over the last couple of years, I hear ya, and no hard feelings - I hope it works out for you moving ahead, and away from the previous chaos that has plagued all of us recently.
A bit of a techie post but it expands a bit more on the discussion about torque. We went back to the dyno yesterday for 4 hours and another 40 pulls. This makes the total number of over 120 pulls as part of the Motec tuning process here. I figured we were pretty much done after last weeks dyno session but Chris scrutinized the logs in details and was not quite happy that everything was perfected. He studied the XR1000 compressor map which convinced him to make a few tweaks. He ran the calculations and plotted the engine output onto the compressor map as can be seen by the straight line with rpm reference points.
He also wanted to finalize the knock sensor tuning and tweak transition fueling when quickly coming on and off the throttle which was not quite to his satisfaction. After about an hour it took him to complete this, he then wanted to set up maps with flat torque curves for maximum drivability. He also wanted to establish a practical and safe tuning limit for 91 octane pump gas. With 91 octane in the tank, Chris went to work doing pulls and scrutinizing timing and any signs of knock which was being monitored live. Amazingly with 91 octane, we managed 981whp / 727 wtq at 1.35bar which is absolutely mind blowing considering pump fuel. We were 20hp away from the 4 digit barrier. Ahh, where is that better Y-pipe? Adding any more boost beyond 1.35bar past 7000 would have the ECU start pulling timing so 981whp was the practical / safe 91 octane limit.
Since we normally run 95 octane at the track, we decided to add 5 gallons of 100 octane to the 5 gallons of 91 that remained in the tank giving us the 95 mix. Chris then added another 0.1 bar boost at 7000 and beyond. Another quick pull and voila, here is 1015whp / 745 wtq at 1.45bar. Mission done. The dyno pull video below really shows how linear and effortless this pull appears.
Finally, we also wanted to establish where we wanted to set the practical torque limit for track use for maximum drivability and engine longevity. When I ran the XR980s, I found the car very easy to drive when torque was limited to 600 ft.lbs as one could aggressively come in with power on turn exit without compromising traction. This level of torque is also very easy on the engine internals. As such, Chris wanted to make a file with a flat 600 ft.lb torque curve which will make for a super linear power delivery. I don't believe a curve can get better than this. This map smoothly ramps up boost with 0.7bar at 4000, 0.8bar at 5000, and just barely kisses 1.1 bar at 7500. The engine pulls strongly to 7800.
Sharp eyes will notice that on the 981 and 1015 graphs, the hp curve peaks and starts to taper off at 7200 and 7100 respectively while on the 858whp dyno the hp curve does not begin to fall off till 7600. What this tells is the fact that the engine begins to fall out of optimum efficiency as the HP approaches 1000. Looking at the logs clearly confirms this and you can also see this at the 7.7K reference dot onto XR1000 compressor map. The Turbine Drive Pressure Ratio (MAP/EMAP) begins to fall below 1.0 earlier in the rpm band as hp increases. At 858whp, the MAP/EMAP does not drop below 1.0 until 7200 while at 1015whp it drops below 1.0 at 6700. As such, in its current form, this engine is essentially at optimal efficiency all the way to redline at up to 875whp or so. By comparison, with the previous XR980s and OEM headers this 1.0 crossover point was at 4500 and down to 0.8 by redline.
One last thing we did was test the Y-pipe by adding a pressure sensor to it at the narrow right leg and could see at close to 0.1 bar pressure drop there. This reading would have been more precise and greater if we had a sensor at the intercooler outlet pre-Y-pipe but we did not have time to do that. With that said, it appears that a larger Y-pipe will free up another 30-40whp or conversely allow for the same output with less boost and less heat, again raising the engine efficiency. Chris and Johnny have already drawn up a Y-pipe on CAD. It will consist of 3.0" legs and a 4.0" collector pre-throttle body. The center section will be CNC machined and the 3.0" legs welded on. This will be a perfect no compromise bolt in swap in place of the OEM Y-pipe.
Finally, we tested the restriction the OEM airbox. We installed a large open tube with a bell mouth as pictured below in place of the airbox for a quick pull to see the restriction. Once the was done, data showed that the while the 100mb airbox restriction was now gone, the 100mb restriction had now shifted further down stream to the intake pipes which before showed a negligible 35mb drop when used in concert with the OEM airbox. To see an overall improvement there, one would have to replace both the airbox and intake pipes together for something better. In a nutshell, this showed is that the airbox in and of itself is pretty darn good when used in concert with the large intake pipes.
Looks like we are going to do one more full day on the dyno for final tweaking and verification of everything and call the engine tuning exercise complete. It's been quite the journey so far. Pretty educational and eye opening. Cheers...
Two things stand out as very impressive about this; one, that torque 'curve' is as flat as a mesa, bravo and well done, should theoretically be better on the gearbox and add to longevity, and two, that intake horn could easily swallow a small child - does it remain, or is the factory box going back in?
That MOTEC has pretty fantastic resolution to be able to see that kind of pressure variation in an intake tract.
Absofreakinmazing! Thank you for all the research testing and sharing with the community. Knowing what y'all know now, how much of this can you reasonably see trickling down into how folks with closer to stock builds/power changing their power upgrade path in terms of improving engine efficiency before adding a ton of boost?
To further clarify, how might one choose their upgrades to get a similar flat torque curve like yours but at a 500-600 wheel hp 400-500 wheel tq?
Pretty much all of this trickles down to more moderate builds. Some things you won't be able to change with the stock ECU like the Variocam profile but others like intake and exhaust restrictions you can. Gains will be there just not to the same degree. All engines are air pumps. The easier you make it for the engine to pump air, the more power it will make with less heat. The other thing I would recommend is to run a larger frame turbochargers especially with the new generation offerings that are on the market today like the XR1000s here. You will get a superb top ends and a mid range that will rival any of the fancy K16 / K24 variants with the byproduct of less heat and stress on the engine.
To get a flat torque curve as you see on our engine will require a turbocharger that has the ability to flow can create power on the top end. Without that, you can't make torque in the upper rpm range. In order to accomplish this, you will need to lower your exhaust back pressure to the greatest extent possible with the use of a low restriction exhaust, larger headers, and larger turbines on. You will never be able to accomplish this with any type of a small frame turbocharger. Running the XR1000s on a largely stock engine will get you close as you can see by the dyno I posted above in #217. With some more hardware updates (better exhaust) and tweaking of the tune, that engine could likely have a flat 500 ft.lb torque curve from 4500 to near 7000. It won't do it at 1.0 to 1.1 bar, but it would get close to that at 1.3 bar with 93 pump and near 700whp on the top end.
Last edited by powdrhound; 10-27-2022 at 12:53 PM.
Two things stand out as very impressive about this; one, that torque 'curve' is as flat as a mesa, bravo and well done, should theoretically be better on the gearbox and add to longevity, and two, that intake horn could easily swallow a small child - does it remain, or is the factory box going back in?
That MOTEC has pretty fantastic resolution to be able to see that kind of pressure variation in an intake tract.
MOTEC can't tell any pressure variations in and of itself. It requires information from external sensors. We have 5 very accurate pressure sensors in the intake track from the airbox all the way into the intake distributor post throttle body. You can see the pre-turbo MAP sensor in the black plastic intake pipe just ahead of the silicone coupler here:
Last edited by powdrhound; 10-27-2022 at 01:18 PM.
But the MOTEC is sampling at 500hz, or more? That will give you some very useful data, especially when using a high quality transducer.
That is correct. With fuel pressure for example, the high sampling rate allows you to actually see the fluctuations in fuel pressure as a result of the injectors pulsing.
Documentation I have says 2000 channels, but I don't know how much memory that would take up to store the logs if you ran all the channels at max rate at the same time so it may make it only possible on paper rather than practical usage.
In practice you only log relevant channels and set relevant sampling rates to those channels. Some stuff only needs 5hz, some stuff needs 500hz, etc.
Appreciated, thought you'd say 24, 48 or 64 - 2000 is remarkable for any DAU; not knowing what the onboard memory is, you are most likely correct that it will not sample 2mhz over all channels.
10-27-2022, 01:14 AM
