Harlan's Impossible Turbo Build 2
10-13-2016, 07:04 PM
#1
Harlan's Impossible Turbo Build 2
Lets skip the foreplay here's what works:
Water injection. (Selectable rate, defaults 25% water to fuel)
Knock Detection. (separate from stock)
PID Boost Control
OBD2 Logging of Fuel Pressure, Boost Pressure, Manifold Pressure,
and associated control parameters.
Teensy 3.1 based piggyback timing control for live tuning.
What hardware I'm running:
Tial 38mm wastegate
Highly modified PTP Turbo kit
8780 SX-E Borg Warner Turbo
60lb EV-14 fuel injector running water. (Used for water injection,
and soaking in water for almost a year with no ill effects.)
Devils Own Water Injection Pump
100% Pure High Quality H2O
S2 Transmission.
Ion sensing for knock detection and PPP detection (Shows where timing is compared to ideal for MBT)
What I'm not running:
Intercooler
Race Gas
Currently developing:
Boost referenced Fuel pump pressure control. (Need to install
hardware, change stock plumbing, and tune PID control.)
Ion sensing for closed loop timing control. (Basic closed loop working in idle, needs refinement.)
Turbo RPM logging (Code written and bench tested waiting on sensor.
Backordered)
Permanent PCB. (Waiting for ion sensing tests to confirm the concept
works)
Possible migration to MS3 with custom software. TBD
Goals:
400HP reliable with relatively flat powerband. ~14PSI maybe less.
40+MPG Cruise.
Here is the current state of the build:
Water injection. (Selectable rate, defaults 25% water to fuel)
Knock Detection. (separate from stock)
PID Boost Control
OBD2 Logging of Fuel Pressure, Boost Pressure, Manifold Pressure,
and associated control parameters.
Teensy 3.1 based piggyback timing control for live tuning.
What hardware I'm running:
Tial 38mm wastegate
Highly modified PTP Turbo kit
8780 SX-E Borg Warner Turbo
60lb EV-14 fuel injector running water. (Used for water injection,
and soaking in water for almost a year with no ill effects.)
Devils Own Water Injection Pump
100% Pure High Quality H2O
S2 Transmission.
Ion sensing for knock detection and PPP detection (Shows where timing is compared to ideal for MBT)
What I'm not running:
Intercooler
Race Gas
Currently developing:
Boost referenced Fuel pump pressure control. (Need to install
hardware, change stock plumbing, and tune PID control.)
Ion sensing for closed loop timing control. (Basic closed loop working in idle, needs refinement.)
Turbo RPM logging (Code written and bench tested waiting on sensor.
Backordered)
Permanent PCB. (Waiting for ion sensing tests to confirm the concept
works)
Possible migration to MS3 with custom software. TBD
Goals:
400HP reliable with relatively flat powerband. ~14PSI maybe less.
40+MPG Cruise.
Here is the current state of the build:
Last edited by Harlan; 12-05-2016 at 07:21 PM. Reason: Update
10-13-2016, 07:49 PM
#2
stopped at 1:02 sec
What are you using to control the wastegate? having it select-able like that is pretty neat
What are you using to control the wastegate? having it select-able like that is pretty neat
10-13-2016, 08:09 PM
#3
I have a push pull wastegate setup with a 4 way solenoid, that is being driven by an arduino/teensy that is interfacing with the CAN network and can accept OBD2 commands through Torque.
I send a 0-255 command to the correct address through my diagnostic port and it sets -15 to 30psi as my boost target. I had it cutting off at 4krpm in the video, so it spooled early but then tried to cut boost after 4k. I fixed that on the next flash to the controller.
I send a 0-255 command to the correct address through my diagnostic port and it sets -15 to 30psi as my boost target. I had it cutting off at 4krpm in the video, so it spooled early but then tried to cut boost after 4k. I fixed that on the next flash to the controller.
10-13-2016, 10:30 PM
#4
that's definitely the coolest feature I've noticed on your build.
How have I missed your videos? I can't wait to see your reaction at 10psi to redline.
10-14-2016, 03:12 AM
#6
I like the non IC idea. Since you apparently, like me, are infected by the non heal-able experimental- nerd- virus, have you looked into spraying HydrogenPeroxide?
If I understood correctly, thing here is that you get the cooling benefits, but you also get an extra Oxygen atom. H2O2 -> H20+0. So you are actually introducing oxygen in liquid form, who are a lot denser than in gas form. A side effect is apparently that exhaust is far cleaner as well. Not that anyone on this forum cares, though...
Hydrogen Peroxide and Energy Independence
If I understood correctly, thing here is that you get the cooling benefits, but you also get an extra Oxygen atom. H2O2 -> H20+0. So you are actually introducing oxygen in liquid form, who are a lot denser than in gas form. A side effect is apparently that exhaust is far cleaner as well. Not that anyone on this forum cares, though...
Hydrogen Peroxide and Energy Independence
10-14-2016, 07:49 AM
#8
I like the non IC idea. Since you apparently, like me, are infected by the non heal-able experimental- nerd- virus, have you looked into spraying HydrogenPeroxide?
If I understood correctly, thing here is that you get the cooling benefits, but you also get an extra Oxygen atom. H2O2 -> H20+0. So you are actually introducing oxygen in liquid form, who are a lot denser than in gas form. A side effect is apparently that exhaust is far cleaner as well. Not that anyone on this forum cares, though...
Hydrogen Peroxide and Energy Independence
If I understood correctly, thing here is that you get the cooling benefits, but you also get an extra Oxygen atom. H2O2 -> H20+0. So you are actually introducing oxygen in liquid form, who are a lot denser than in gas form. A side effect is apparently that exhaust is far cleaner as well. Not that anyone on this forum cares, though...
Hydrogen Peroxide and Energy Independence
Hydrogen to rotaries is what monatomic gold was to the annunaki.
10-14-2016, 02:47 PM
#10
If I understood correctly, thing here is that you get the cooling benefits, but you also get an extra Oxygen atom. H2O2 -> H20+0. So you are actually introducing oxygen in liquid form, who are a lot denser than in gas form. A side effect is apparently that exhaust is far cleaner as well. Not that anyone on this forum cares, though...
Currently spooling at ~2700RPM and hitting boost hard by 3200RPM. I can hit 8psi well below 3500. It's hitting too strong, and too low for my preference. I expect to be hitting boost about 4000-4500 with this setup, and hitting it hard. We will see.
10-14-2016, 08:11 PM
#12
I was hitting 8psi <3500 in second gear. It was a rip roaring good time, and somewhat terrifying because the first time was from a boost line failure. That was before I put the SS braided brake lines as leaders to the WG. No problems since.
My biggest problem (?) was having no IC means when boost comes on fast, add to that the WG holding shut until it hits setpoint, and you go from 0psi to setpoint in an instant. It just wasn't very street-able. It was like having an on/off switch at the bottom of the pedal, and above 4psi it was fine for a cheap thrill, but no fun around town.
I'm not exactly sure how low it was hitting boost, I've always had a rpm and/or MAF threshold set to trigger the WG closed and WI on. This was to prevent WI at idle/cruise and prevent boost without WI. As soon as it hits the threshold it spools. I wasn't going to do a 5th/6th gear pull from 2000RPM either because that's silly and bad for the engine.
I was able to dial in PWM on the boost controller to make it a lot more gentle, but after doing the math on match bot it just made sense to get a bigger turbo. I should be able to get into boost between 4k and 4500, I don't have the math I used in front of me, but it should be in that ball park. If its at 5k I'll still be happy.
The small turbo would have been limited to about 8psi, (for around 20psi back pressure) beyond that it would have gotten dicey. I should be able to make 14psi with this new configuration and maintain under 20psi redline.
The turbo I had was a t4 .64AR 68mm turbine and a .70ar 82mm compressor china charger, I'm moving up to a 87mm compressor and a .88AR 80mm turbine. Yes it is a huge step up, but I believe I can spool it low enough for my liking, if not I'll find something a bit in between. I'm still feeling this out, but I got enough of a feel to commit to a turbo. If I have to try another after this so be it. I did have some hesitation with going this big on the turbine housing, but hey you only live once.
Oh and I have to modify the battery tray to make the turbo fit. But the hood does close, so there is that.
My biggest problem (?) was having no IC means when boost comes on fast, add to that the WG holding shut until it hits setpoint, and you go from 0psi to setpoint in an instant. It just wasn't very street-able. It was like having an on/off switch at the bottom of the pedal, and above 4psi it was fine for a cheap thrill, but no fun around town.
I'm not exactly sure how low it was hitting boost, I've always had a rpm and/or MAF threshold set to trigger the WG closed and WI on. This was to prevent WI at idle/cruise and prevent boost without WI. As soon as it hits the threshold it spools. I wasn't going to do a 5th/6th gear pull from 2000RPM either because that's silly and bad for the engine.
I was able to dial in PWM on the boost controller to make it a lot more gentle, but after doing the math on match bot it just made sense to get a bigger turbo. I should be able to get into boost between 4k and 4500, I don't have the math I used in front of me, but it should be in that ball park. If its at 5k I'll still be happy.
The small turbo would have been limited to about 8psi, (for around 20psi back pressure) beyond that it would have gotten dicey. I should be able to make 14psi with this new configuration and maintain under 20psi redline.
The turbo I had was a t4 .64AR 68mm turbine and a .70ar 82mm compressor china charger, I'm moving up to a 87mm compressor and a .88AR 80mm turbine. Yes it is a huge step up, but I believe I can spool it low enough for my liking, if not I'll find something a bit in between. I'm still feeling this out, but I got enough of a feel to commit to a turbo. If I have to try another after this so be it. I did have some hesitation with going this big on the turbine housing, but hey you only live once.
Oh and I have to modify the battery tray to make the turbo fit. But the hood does close, so there is that.
10-16-2016, 08:24 PM
#17
Sorry, I was pretty vague. I meant it's hitting the 4psi target at those numbers. Here is a graph of RPM vs Manifold pressure in 4th. I haven't found a good apples to apples run on the old turbocharger in my logs, but it was hitting 4psi WOT about 3000-3500 RPM in 2/3rd. It may have even been earlier than that, but I never looked low enough to tell.
10-17-2016, 06:42 AM
#20
10-17-2016, 01:34 PM
#21
Something that will improve it a little is to hold the ssv closed till 4000rpm . Swapping the letters on the side from sxe to EFR would be my pick though.
Last edited by Brettus; 10-17-2016 at 02:01 PM.
10-17-2016, 05:56 PM
#22
Yeah, it is a bit sluggish. I want to try a few things before I decide.
I need to stop the wastegate from opening at low RPM so it will be spinning faster in cruise. Later I will code it to do this automatically on a delay.
I need to see what 6psi looks like.
And I need to try retarding timing while spooling and light a bigger fire under the snail.
What I like is the engine no longer feels restricted at higher RPMs so everything is breathing easier. If I can get a better spool in 2nd/3rd I'll keep it.
I need to stop the wastegate from opening at low RPM so it will be spinning faster in cruise. Later I will code it to do this automatically on a delay.
I need to see what 6psi looks like.
And I need to try retarding timing while spooling and light a bigger fire under the snail.
What I like is the engine no longer feels restricted at higher RPMs so everything is breathing easier. If I can get a better spool in 2nd/3rd I'll keep it.
10-17-2016, 08:04 PM
#24
Did the same run today, only paid more attention to what gear I was in.
When I said before it was hitting 4psi at 4k in 4th, it was really third. It was hitting 4psi at 5k in 1st/2nd.
After tweaking the code to keep the WG closed under low load/rpm it now hits 4psi at about 4500 in 2nd. Maybe changed 3rd by 250RPM if that, probably because it has enough time to spin up in 3rd.
It did hit 6psi at 5500 in 2nd, and at 4000rpm in 3rd So it's really not that bad right now. I'm not to the point of throwing it out and trying again, but might try a smaller turbine housing if push comes to shove.
It's in line with my original estimates, and although I lost the low end I had before the top end is wide open. Might be able to tune it down another 250-500 rpm or so by retarding timing. We will see.
The EFRs have nice billet compressor wheels, which are the same as on the SX-Es, dunno about the hot end though. There may be some magic with their internal BOV/recirc, but right now it's not worth the money to me. Remind me in 6 months.
When I said before it was hitting 4psi at 4k in 4th, it was really third. It was hitting 4psi at 5k in 1st/2nd.
After tweaking the code to keep the WG closed under low load/rpm it now hits 4psi at about 4500 in 2nd. Maybe changed 3rd by 250RPM if that, probably because it has enough time to spin up in 3rd.
It did hit 6psi at 5500 in 2nd, and at 4000rpm in 3rd So it's really not that bad right now. I'm not to the point of throwing it out and trying again, but might try a smaller turbine housing if push comes to shove.
It's in line with my original estimates, and although I lost the low end I had before the top end is wide open. Might be able to tune it down another 250-500 rpm or so by retarding timing. We will see.
The EFRs have nice billet compressor wheels, which are the same as on the SX-Es, dunno about the hot end though. There may be some magic with their internal BOV/recirc, but right now it's not worth the money to me. Remind me in 6 months.