The Forced Induction VR6 thread

[craziscot 0]

Just what I was looking for Alex!! :D

 

On top of the increased boost low down and improved performance due to its gearing, it uses the standard belt tensioner which greatly reduces belt slip and all the lovely issues that go along with that on the VF system. Belt is good for 30K miles.

 

The charger is good for 450 bhp easily achievable on a VR with lowered compression.

[StormchargedVR6 0]

could someone tell why the power of the Rotrex is so good but torque so low. Storm developments 1.8 20vt is 325 bhp & 254lbft

 

I've seen this on few Rotrex chargers :?

[Kevin Bacon 5]

Raj at NSRacing can get you a proper VF engineering 12psi pulley mate, £45. With that I actually got slightly more, 12.5psi at 6900rpm.

 

Rotrexes - They are basically gear driven turbos. They can hit 100,000rpm just like a turbo. The V9 can't spin that fast unforunately. The V9 is a true centrifugal supercharger, which doesn't have as much in common with a turbo as the Rotrex does.

 

The reason some torque figures are low on Rotrex based engines is mainly due to the type of engine they're fitted to. For example, the Aerial Atom has well over 300hp on the VTEC engine, but the torque hardly increases at all, but it's only running something like 6psi and the compression ratio is too high to go mad with boost.

 

On a VR6, the power and torque would be quite similar as that's how it is as standard.....bore and stroke, cam lift and duration, compression ration, intake and exhaust pulse tuning and mapping all govern that.

 

Also depends on the size and trim of Rotrex you use too. The bigger ones will give plenty of torque.

[StormchargedVR6 0]

spoke to Raj friday , but hes out of them. His brother nick in the states is out to. Spoke to chris at C2 he got one.

 

Kev Thanks for clearing up the rotrex power thing

[Kevin Bacon 5]

Engine Management for FI

 

As promised, here is my P8PRo mapping to help anyone out considering the same ECU.

 

The map is set for - 2937cc 8.5:1 Compression, Delphi 42lb (440cc) 'disc' injectors @ 3bar, NGK BKR7E plugs gapped to .035", MSD coils @ 2.5ms and Innovate LC-1 Wideband controller. This will get you 90% there, with some basic tweaks required for your own engine setup. The MAP is good for 300kpa (29psi) of boost.

 

I've found the Delphi injectors and BKR7E plugs to give the best idle and part throttle qualities. Don't waste your money on expensive plugs, the VR T runs best with cheap plugs, especially with the MSDs. Audi S3 plugs don't gap well at all with the MSDs, they clearly don't conduct aswell as the cheap ones.

 

I've got on well with this map and it drives nicely. It's a map I've done myself and am happy to share it.

 

It's a TPS v RPM map, but you can use MAP v RPM if you want to, which is easier and sometimes more responsive, but it can cause bucking in the lower revs. TPS v RPM is smoother for a daily runner.

 

This is just the basics to get you running, the P8 can do a lot more, but some of it isn't necessary.

 

General Engine settings - These are the core essentials.

 

 

Main fuel map - If you're using different size injectors, adjust the map by percentage of flow rate.

 

 

Timing map - If you've got a short runner and low compression, you can be generous with the timing in the midrange, which will give good off boost torque.

 

 

Throttle stops - This is where you tell the DTA your min and max throttle positions, which is very important. You must reset this if you adjust the throttle cable or throttle stop.

 

 

Start fuelling - Pretty arbitary numbers really, just chucks a load of motion potion in to get the motor started.

 

 

Water temp compensation - Fairly self explanatory. In here you can give the motor a little fuel and timing against temperature.

 

 

Intake air temp compensation - Same again.

 

 

Throttle pump effect - Just squirts in a little dash of fuel when you press the pedal hard (configurable pedal speed)

 

 

Lambda settings - I run closed loop all the way through the revs. Be sure your wideband is reliable before doing this! Normally you would go open loop at 4500rpm and above 85% throttle and fall back to the base map for safety, but full closed loop will work well with the adaptive tuning to adjust the base fuel map, then you can use open loop settings after it's adjusted the base map. This map take care of idle a/f and over run for you too, which switches the injectors off at throttle position 1. Good for economy and it's smooth too, no annoying popping and farting at all.

 

 

Lambda Target map - This is where you tell the DTA to keep your A/F at any given rpm or TP. This is a lean map and works nicely, gives quite good economy. 30mpg easily achievable off boost. I find VRs idle best at 13.8 AF. Past 14 they get a bit lumpy and grumpy due to evap down the rear bank's intake tracts.

 

 

Fuel Correction - If you ticked the Generate Fuel correction map option, it will offer you some adjustments to make to the base fuel map, which you can choose to apply or ignore. I would ignore the recommended changes in the low rpm cells and focus on the mid / high ones. Part throttle tuning is a pig. Lambda probes are cold and therefore unresponsive at idle and low rpms, so if it drives OK on part throttle, don't apply the changes!! Zero them out and apply the mid/high ones only. The correction table is a rolling update and the numbers should become very small after a few applys.

 

 

Pressure compensation - This is an important one as it governs your boost fuelling and timing. These settings work best I've found, but basically, the better you get the base maps, the less you'll have to do in here.

 

 

Injector Phasing - If you want to trim individual cylinders, you can do so here, but you'll need to run the injection as sequential, also set in here. This is great for compensating for longer inlet tracts on one bank for example. I'm not using this at the moment, but will be soon.

 

 

Turbo control - You can use a standard N75 boost valve to control boost pressure for you, again, not configured yet, but it's in progress! In here you can set boost against rpm if you like. PWM control is preferable to PID, given the choice, which you are here thankfully!

 

 

The only thing I haven't gone into yet is idle control, but it has a full PWM 20 x 14 map for that. I'm just waiting on this because I'm expecting a firm/soft ware update from DTA to allow idling speed to be set with timing instead of a valve, which is MUCH simpler, more reliable and neater.

 

So there's no excuse for not getting a standalone now, all the hard bits have been done for you 8)

[alex303 0]

The charger is good for 450 bhp easily achievable on a VR with lowered compression.

 

Ah but your getting the 'full fat' version 30-94 mate (lucky devil), as opposed to my 'semi-skimmed' 84 :( , think you'll definitely wipe the flr with me in a straight line! :) Hmmm wonder if you can upgrade the internals on an 84, now thats a thought!

[CrazyDave 0]

Excellent stuff Kev, not sure I'm brave enough to post my settings yet!

 

Wouldn't want anybody to kill an engine by using rough mapping from me. :)

 

You seem to have several more degrees of ignition advance than me across the board, might give that a go on the MegaSquirt.

[Kevin Bacon 5]

You've got lots of favourable comments in your thread from people who've been in your car, so I don't think your mapping is that bad! :lol:

 

You can run more timing with the MSDs and I also have a lower compression ratio than you ;-) That said, it is running a fair bit of timing, which is fine so long as there's no pinking.

[CrazyDave 0]

I must be getting used to the extra go :lol: !

 

Not good cos I'll be tempted to go for a bit more boost!

 

I'm still not 100% happy with the accel enrichment, needs a bit of tweeking to prevent a little slight lean tip-in I'm getting.

[Kevin Bacon 5]

You do and will!

 

The boost is there on tap, go for it :-)

 

Yeah the momentary lean-ness on part throttle is fun and games to map and it can't be done on a dyno. I had that issue too. Got rid of it by running both the ECU and Lambda refresh rates as slow as possible (1/3 sec), driving to the cell with closed loop disabled and making sure the cell isn't too lean and reducing the lambda adjustment to 10% + / -

 

The throttle pump effect can work too, but the danger with that is the lambda will see a momentary richness and try and pull it back....which might be what's happening on yours.

 

What does your live mapping screen say when you drive up to and through that site?

[CrazyDave 0]

The closed loop AFR is set to be fairly slow, so I don't think its that (and it still does it when I turn it off anyway).

 

So it looks like its when you change throttle state, when it's steady at the site it's fine, but if you open quickly it dips lean, and when you close quickly it dips rich. Think it's fine tuning of the enrichment (or EAE as it's called in Squirt). It's happening before the map sensor sees a change so it could be the map sensor filter is set to high, but on boost the sensor is a bit 'frisky', so maybe not.

 

Minor stuff though, just me being fussy.

[Kevin Bacon 5]

Sounds to me like injector latency. A/F response is a lot faster than injector response, hence the apparent leanness.

 

Can you compensate for injector dead time in the Squirt?

[CrazyDave 0]

OK, I've had a quick read of the DTA manual to covert from DTA'isms to MS'isms, and MS calls this initial injector opening time.

 

I haven't played with this much and it's currently set at 1ms. I'll try tweeking this down a little as I do have modern injectors that probably open very quickly.

 

The reason I think it's accel enrichment is because my injector PW doesn't increase enough when I open the throttle quickly, so thats causing it to go lean momentarily until the map catches it. It's tricky to tune unless you use datalogging to see whats happening. It's all, do a run, check the results, adjust etc etc. Just takes time to get just right.

 

Looking forward to Inters, standing quarter here we come! Last year, the standard car (and I mean standard, exhaust, paper filter the lot) did 0-60 in 7.32 and 15.706 for 1/4mile. Not bad considering I had a full tank of fuel!

[Kevin Bacon 5]

Yeah I reckon it might be cured with some throttle enrichment.

 

It certainly is a time consuming process tuning the low rpm areas! God knows how much petrol I've gotten through tuning the DTA, but I've found even fuel temperature affects things. Ever notice how on a hot day with a near empty tank, it feels OK ish, but after a fill up with cold, fresh V power, she flies!

 

Is there such a thing as a petrol cooler? I think it would make a huge difference personally.

 

Yeah I always drive with a full tank to ensure my timing settings don't cause pinking as the weight increases.

 

Good luck getting the power down at inters :lol:

 

I was spinning both wheels in the dry in 2nd this morning....I think I need to get my geometry checked and the WB bushes replaced as they looked a little cracked last time I looked.

[CrazyDave 0]

Yeh, fuel temperature sure does influence performance, pretty much all the test cells we do have fuel temp conditioning to maintain a constant temperature for consistant results (diesels suffer with this much more though!). Trouble is you need a chilled water plant to make it work! Maybe fit air conditioning but use it to chill the engine air / fuel instead! :)

 

Weather for Inters on Saturday is looking wet.... maybe I'll see how many miles of wheel spin I can clock up in a quarter mile! :lol:

[Kevin Bacon 5]

Yeah I was thinking of those really thin A/C rads Audis have in front of the main radiator.....but for cooling petrol :lol: It makes such a massive difference, but I suppose OEs would argue that a constant flow of fuel through the circuit should keep the rail temp constant, but we know that's not true when you're sat in traffic for ages 8) Audi got round that in the 200 turbo by fan cooling the rail :shock:

 

LOL, 1st, 2nd and 3rd won't be much use with an open diff, full throttle in the rain.....especially as the 1/4 tarmac will be a cocktail of water, oil, petrol, diesel and rubber 'marbles'.

 

Been refining my fuelling setup a bit due to petrol smells on boost recently.

Until last night, I was using a modified FSE reg, but I don't think it had the guts for turbo use. So I dug the SX out of the cupboard and put him back where he belongs, after first modifying it to work as a return loop regulator.

 

Yet again, it's improvements, improvements, improvements!

 

I took it off originally because I thought a 1500hp rated fuel reg was far too overkill for my application, but seems not. It also holds 25psi pressure to improve hot starting response. Fires instantly. Idles superbly and throttle response is much better.

 

This is an old pic, as I've now got a rail return, but the SX is a serious bit of kit..... recommended.....

 

[Simon69 0]

Hi there,

 

Dave have you posted any more information about your Megasquirt setup in the forum? I've read through all 8 pages of this and also done the search thing but most results seem to kick me back here.

 

I would be interested in hearing about your megasquirt setup in more detail.

 

I have PM'd you too, I do have reasonable expierience with ECU calibration...

 

btw, on the Celica WRC they ran the fuel through a small heat exchanger to help control det (detonation / pre-ignition) by keeping the fuel temp more or less constant, some turbo circuit racers do it too.

 

Cheers

 

Simon

[CrazyDave 0]

Simon,

It's not posted yet, but if your looking for a basic setup for a VR6T then I can send it to you. I'm not completly happy with it yet but it's a real good start and it's been running on my car for about 2500 miles now. I'm using an MS II processor with v1.0.2 MSII_Extra code.

 

If you've been through my gallery then you'll have the basic spec of the engine, turbo etc that it's setup for, but just in case:

 

2.9 VR6 with stainless spacer and MKIV multi layered head gasket.

 

GT3076R turbo with 8psi wastegate.

 

PWR chargecooler 10"x4" with 10" front mounted aux rad and fan.

 

Cams are standard, as are pistons, rods etc.

 

ARP head and rod bolts.

 

Custom fabricated short manifold.

 

None of the original car wiring has been cut at all, and an old Motronic box serves as a termination point for the MegaSquirt looms. MS box is in the glove compartment.

 

I'll post my latest MSQ stuff in a day or so, getting ready for Inters at the minute. :D

[CrazyDave 0]

Well not quite the standing quarter result I was looking for :lol: .

 

I seem to be rather down on power? Which kind of explains why I'd been getting used to it. :(

 

Best quarter time was 15.054sec not great considering, and looking at the acceleration curve from the radar trap it would appear I'm getting good grip, even if first gear doesn't last very long. Just not enough go!

 

Decision made then, time to book some time on a rolling road to get the top end power sorted.

 

Here is a link to my MegaSquirt MSQ file

http://www.d2designs.vispa.com/VR6%20Turbo%20A.msq

 

Goes without say really but I'll say it anyway 'Use this information at your own risk!'.

 

Target AFR table for closed loop control of fuel.

 

 

Ignition Advance.

 

 

Fuel Table.

[Kevin Bacon 5]

Nice one Dave, cheers for the info.

 

Nice to see your fuel scale is in % instead of ms! Much nicer 8)

 

I think a little more fuel in the boost areas might help with the power. What are your EGTs on boost?

[Simon69 0]

Hi Dave,

 

Thanks for that, great to see you posting them up in % too!

 

For what it's worth I would agree that it looks a little lean in the load areas, you might want to try a little more fuel when coming onto and in boost (i.e. more than theoretical "best power" (12.7:1)) in conjunction with a bit more ignition all over.

 

Though this moves you away from the theoretical best position I have found that running the ign closer to det (or even slightly in det for LIGHT load) and then overfueling the load areas to "put out the fire" gives better turbo spool and better part throttle economy and responce... "Kind of have your cake an eat it" :)

 

This is mostly based on Subaru's of course and they are all alloy engines that just love to die horribly at the slightest sign of det ;) with your big tough iron block you can run close or in slight det without so much worry, or course, how much det is too much det is the $1,000,000 question lol...

 

I'd be curious to know the EGT's too, also I'm assuming you've checked that the engine is hitting the afr and ign targets in the map?

 

Cheers

 

Simon

[CrazyDave 0]

Good info guys!

 

Have a look at the logging from one of my runs on Saturday.

 

 

Top trace RPM, MAP, CLT (coolant)

 

Middle MAT (manifold air temp), AFR, PW (injectors in ms)

 

Bottom TP (throttle position %), SparkAdv (degrees), WallFuel (accel enrichment correction) this is way off still and needs work.

 

Good call from both of you, boost is a little lean so I'm going to try a little more fuel in those areas. And if you haven't spotted already my fuel table only goes to 150% so I'm right on the edge of it, needs rescaling to cover to 200kPa like the ignition.

 

Quite a lot of oscillation on the MAP sensor when in boost and it only just hits 150kPa max, most readings are 130-140 ish? But I guess the signal is showing some aliasing in the log file.

 

EGT isn't part of this trace unfortunately due to this being the MS-II extra code, so I have a separate data logger for that which I didn't have hooked up for this run. But when it has been, I see 4-600 ish degrees C at cruise and 7-800 when accelerating WOT. This seems a little cool? Makes me think the timing is getting a little bit to advanced?

 

Ignition advance looks very flat doesn't it? It's pretty much all in at this engine speed.

[KADVR6 0]

is that english you just typed dave, or is it klingon :lol: some of you guys can really confuse a simpleton like me, very very easily :lol:

[Simon69 0]

Interesting data, my thoughts,

 

The afr trace (well all the traces) seems very jumpy, what frequency are you logging at?

Also the afr log seems to be quite different to the "afr request" in the table?

The ign seems almost "fixed" at circa 27 deg right through the rev range?

The intake temp seems high, were is the sensor?

Do you have a proper cold air intake on the car?

 

On the EGT is would be my opinion (and I'd welcome Cheesewire (Kev?) input here) that your 7-800 is on the low side because your not yet "pushing" things very much (i.e. quite rich and very conservative ign settings). for info, I would expect to see 900+ on a scooby with a sti turbo (over 1000 overspeeding the stock unit ;) ) that would be with 2.5bar absolute in 4th from 3k to redline, ign about 15deg at 4k raising to 33 ish at 8k. There would be some det with this but not too much, oh an thats on californian gas which is 91Ron+Mon/2, roughly the same as our 95ron pump gas.

 

Which brings me to the last Q lol what do you run it on? regular 95 or Shell Optimax? We did quite a lot of testing on fuel and Optimax always came out on top for a turbo motor

 

I'd need to run it past Vince but my initial feeling is you need a sh*tload more advance (unless these units won't take ign advance for some reason)

 

edit, sorry, one more, do you use det cans when your tuning?

 

Cheers

 

Simon

[CrazyDave 0]

MS logs at about 20Hz and I try to keep the software lag filters as low a poss, to give good response.

 

The AFR closed loop isn't active in boost, and I'd turned it off for these runs anyway (standing 1/4 mile at GTI International by the way).

 

The igniton must be down to my map settings, which I don't think are correct.

 

The high intake temp is due to the sitting at idle for about 25 minutes with the engine on and off waiting for the run. But you have a point with the sensor it could do with being placed into the air stream (it's in the base of the plenum chamber at the moment) and changing for an open thermister bead type, which would be much better for a boosted engine. The air is colder than the trace, I have a thermocouple in the boost pipe post charge cooler and it's a good 20 degrees lower than the MAT. So another problem area.

 

Cold air feed is in the front off side bumper via the old hole for the carbon filter.

 

Nail on the head with the timing I think. My forum name may be Crazy but I'm quite the opposite when it comes to it! I just don't have any information on how far to push the timing. I have a knock sensor connected to the MS so I was going to use that to get the timing sorted. Timing figures so far are based on VAG-COM datalogging of the standard car with Motronic ecu. This seems to hold steady at about 24-26 degrees from 3500k to 6000k.

 

Minimum super unleaded 97 (Esso) and usually 99 V-Power.

 

Thanks for the info, much appreciated!

 

 

Karl, it's back to school time again! Keep reading though, most of it is still a dark art for me! Hopefully the light will come soon. :)