So my Fiero will have a Megasquirted iron-head turbo V6 engine... It will have a knock sensor (not sure how well it will work).
Last week, somewhat by accident, I stumbled upon a dynapack dyno, in an auto modification shop situated about a 20 minute drive from my home. Sweet!
Right now, I have a spark map which is 100% guessed. So on the dynapack (which is a brake-type dyno) I should be able to hit each cell on my timing table, and tweak the spark in each cell in order to find MBT, or perhaps find the knock limit.
On the street, I plan on running 91 octane pump gas, because that is the highest octane which is readily available.
So I was thinking, should I be tuning using higher-octane race fuel? This would allow me to find (and even overshoot) MBT, without needing to worry about knock. Then, I could find the knock limit (without the dyno) separately on 90 octane pump gas (50-50 mixture of 89 and 91). Divide and conquer a complex problem into two bite-sized chunks, right?
Using race fuel on the dyno could help prevent me from blowing up my engine during tuning.
However, this strategy will only work if the racing fuel has a similar burn rate to pump gas. If the burn rate differs too much, then my timing table will not be optimized for 91 octane.
Any thoughts on this practice?
Seems like i can get VP Racing Fuels nearby, so I'll be asking VP about their thoughts.
Here is my guessed spark table... If I'm going to do this build right, I should populate the table correctly one day!
There's a Petro-Canada 94 station near my home. Thanks for reminding me!
I forgot about the existence of these sporadically-located 94 pumps; since I want my Fiero to be capable of long road trips with no special fuel needs (which means 91 octane), I have become blind to 94 octane pumps.
On top of that, it should be pretty similar to pump gas, since it is pump gas!
But now I realized a fatal flaw in my logic: if I use high-octane fuel on the dyno to not blow up the engine, I'll just be transferring the problem of finding the knock limit to the street when I switch to 91 octane.
How was your 3.4 supercharged tuned? What work did Wester's Garage perform?
Question to all: For the higher-load cells (where the knock limit will be reached prior to MBT), I am not eager to advance the timing in each cell until there is knock, and then back off. Is there a better way? Is finding the knock limit for each cell abusive for an engine, or is this standard dyno practice?
I have seen timing tables that only compare RPM, MAP readings and the timing degrees so I cannot offer any advise on your chart. What I will say is that you must be very careful with timing advance on a turbo engine. It must be slowly retarded as the boost increases. What we have used here is a split timing chart that sets the timing advance in both vacuum and boost modes, at all RPM and MAP readings at each RPM. I can try to pull out the chart and make a copy but IIRC timing at 10 psi of boost is only like 5* advance on top of the 10* initial timing. I would highly recommend using a knock sensor too.. As for gasoline, I would recommend using 93 octane pure gas. If this is just a show/hobby car there are a number of quality octane boosters that work real well and there is some garbage on the market as well. The quality products are costly.
------------------ " THE BLACK PARALYZER" -87GT 3800SC Series III engine, custom ZZP /Frozen Boost Intercooler setup, 3.4" Pulley, Northstar TB, LS1 MAF, 3" Spintech/Hedman Exhaust, Autolite 104's, MSD wires, Custom CAI, 4T65eHD w. custom axles, HP Tuners VCM Suite. "THE COLUSSUS" 87GT - ALL OUT 3.4L Turbocharged engine, Garrett Hybrid Turbo, MSD ign., modified TH125H " ON THE LOOSE WITHOUT THE JUICE "
I would do your Dyno with whatever fuel you plan to run on the street. After you have it dialed in, then you can go back to the dyno with a tank of race gas, boost your timing and see what she'll do.
I have seen timing tables that only compare RPM, MAP readings and the timing degrees so I cannot offer any advise on your chart. What I will say is that you must be very careful with timing advance on a turbo engine. It must be slowly retarded as the boost increases. What we have used here is a split timing chart that sets the timing advance in both vacuum and boost modes, at all RPM and MAP readings at each RPM. I can try to pull out the chart and make a copy but IIRC timing at 10 psi of boost is only like 5* advance on top of the 10* initial timing. I would highly recommend using a knock sensor too.. As for gasoline, I would recommend using 93 octane pure gas. If this is just a show/hobby car there are a number of quality octane boosters that work real well and there is some garbage on the market as well. The quality products are costly.
Sorry, I forgot to label the axis of my table correctly. ignload % here is is the same thing as MAP kPa.
I bought a copy of High Peformance Fieros, by Robert Wagoner, because it contains a spark table for a turbo Fiero.
However, for best results, the spark should be fine-tuned for the engine. My mods won't be the same as theirs, and they blew up their engine... So I don't want to use a canned tune; thus why I'm trying to figure out the best way to custom-tune my engine.
I'm planning on driving this car about 5000 miles a year, so it needs to drink the readily available 91 octane here! When I go to higher boost later (provided I don't kill the engine), I will add water/alcohol injection (going to try without intercooling).
As mentioned, I have a knock sensor, but I might buy/make det cans for the dyno.
quote
Originally posted by fierosound:
They ran the engine under load on the dyno and changed the fuel tables in the 7170 ECM's chip. They left timing tables all stock.
More recently, I've installed a 7165 ECM with the 12P BIN and it's been tuned using TunerPro RT.
I guess if I can find the spark tables in those computers, then that gives me a point of reference.
quote
Originally posted by edfiero:
I would do your Dyno with whatever fuel you plan to run on the street. After you have it dialed in, then you can go back to the dyno with a tank of race gas, boost your timing and see what she'll do.
I think I'll just dyno with 91 octane then. I'll probably spend $500 or so at the dyno for a 91 tune, so I'm not gonna drop more $$$ for a race fuel tune I can't even (well, rarely) use. Could be more money that that, who am I kidding...
[This message has been edited by pmbrunelle (edited 05-12-2017).]
Does mega squirt have a table for knock reduction? If it is similar to the ECM that controls my turbo 2.2 ecotec (stock 2004 cavalier with HP tuners), then you can tune that table to gradually reduce spark advance as knock is detected. These computers monitor all functions thousands of times per second so the knock gets detected and spark will get retarded before any damage can be done. I am barely a novice tuner but I have learned a few things: 1: High intake temperatures are your enemy. If you are not going to use an intercooler then you really need a water meth setup. 2: If you only use water meth injection your intake air temps will look high, even though the charge is getting cooled. So you have to take a bit of a leap of faith and tune your KR to reflect this. 3: There usually is about 5 tables that need to be adjusted to make everything work together. I do not know what MS gives you to work with, but look at all of them.I got a lot of my tuning off of the HP tuners website, sometimes I could find someone with a similar setup to mine and I copied their tables. I ran in to a problem that turned out to have nothing to do with my tuning. I was getting a huge drop off in power above 6300 RPM. After trying everything I could think of tuning wise, I realized that my turbo was being starved by the 2" exhaust system I was using. I upped it to a 2.5" system and now I can get power past 7000 RPM.
The MS ECUs are really crude; way behind any ECU from a fuel-injected production car, let alone one from the 2000s. However, if you don't have to meet OEM-grade emissions/driveability/fuel economy requirements, it's great to tinker with.
The simplicity of MS is its greatest characteristic. There is no hidden function that will unknowingly mess with your timing based on IAT. Timing can be based solely on a single 2D RPM vs MAP 16x16 table.
If you want, you can enable IAT-based timing retard, but the key is that it does exactly what you dictate, and no more. This ECU is simple enough to fully understand its operation.
The knock sensor setup is dirt simple. If knock is detected, retard is immediately applied to the timing from the main spark table. After the knock event, timing slowly creeps up again (user-defined parameters) until there is no more retard, and the timing advance from the main spark table is used as-is.
I'll be tuning the car as if it had no knock sensor; it's really there to act as a safety net. I don't want to rely on it on a regular basis.
I'm intentionally putting the sensor before the turbo, so it won't see the temperature rise due to compression of the air. Even the open-element GM IAT sensor takes a few seconds to react to temperature increases, as per a bench test I did here: http://www.fieromontreal.co...25.msg17530#msg17530
Since it takes a few seconds to react, the temperature reading would be wrong for the first few seconds in boost. Therefore, rather than have post-compressor sensor that is inaccurate on transients, I decided I would prefer a pre-compressor sensor that always reads the ambient temperature accurately, all the time.
Charge temps don't get unexpectedly hot for no reason. Charge temps get too hot if the ambient air gets too hot. Therefore, if you wanted to implement IAT-based retard, you could do it based on ambient air temperature. It's just that instead of retarding at 80°C (if using charge temp), you might retard at 30°C (if using ambient temp) instead.
But for now I'll be doing the low boost thing (as in 6 psi) with no intercooler, no water/methanol injection, and get the car on the road this way.
I'll add water/methanol injection later, as a "Stage 2" project. My plan is to inject the water/methanol into the intake ports using the IAC passageways of the Fiero 2.8 manifold. So for now, I need to start a chemical compatibility test. I will half-submerge the manifold in a bucket of -35°C windshield washer fluid (my planned boost juice). Next summer, after a year of exposure, if the aluminium manifold hasn't been too attacked by the methanol, then I will go ahead with the water/methanol "Stage 2" project, and crank up the boost (Sunbird boost control solenoid).
The obvious answer to your main question in the OP (obvious to me at least) is You should tune your car on whatever fuel you plan to use. If you tune it on pump gas the tune will be more aggressive than it should be for pump gas.
With the Honda method, you don't need to change the MBT table due to octane changes. However, you might change the knock limit table depending on fuel octane.
With the MS (without mods at least), you have a single spark table, so you need to make it follow the green trace.
I was wondering if there was a benefit to finding MBT for every cell (writing down the value on a paper) using race fuel, before proceeding to finding the knock limit for every cell. I would then take the most retarded of the two values, and then load that into the MS. This is effectively the same as what the Honda ECU does in real-time, but using a pen/paper method.
I am somewhat convinced that doing exhaustive timing sweeps for each cell is perhaps too abusive if I want my engine to survive tuning. For part-throttle, this is probably OK though.
