Hi guys,

In the August issue of Grassroots Motorsports Magazine we have a 5 to 6 page write up outlining the installation of the a Megasquirt system into my 84 RX-7 GSL-SE. You can see the mods in my sig. Jerry and the guys from DIYAutoTune along with rx7club.com megasquirt guru Ken Culver (muythaibxr) installed the system over a two day period at the Mitty.

I wanted to personally thank Ken for coming down from Maryland to help out and for not getting too upset when Per tried to kill him with a tree.

We saw a gain of almost twenty peak HP. The car is currently running great aside from an intermittent stumble from when the second injectors stage. We believe the timing is off due to having an incorrect pulley on there. With corrected timing and an S5 intake swap we are hoping to see another ten horsepower.

Look for all the details in the August issue which should be on newstands soon and in subscribers hands right about now. You should be able to find the magazine at Barnes & Noble, Borders, or Books a Million.

Or you can hop on our website and get a free trial issue. www.grassrootsmotorsports.com

Let me know if you have any questions.

Thanks,

Greg

 

Haha, I wasn't going to mention the tree incident. In any case, it was fun, so no problem. Glad I could help.

I think we could fix that with more time spent on the staging settings. I'm also working on a code mode that will change the way staging works so that the primary pulse-width doesn't drop as fast (which is what causes the slight lean spot).

The way staging is set up right now on your car was to stage at a low RPM where a slight lean spot won't hurt anything. We had it set up earlier that day when we were tuning it so there was no stutter (using duty-based staging), but the AFR went a little lean while it staged, and it was staging right around the same point your engine started to wake up. There was a definite slight dip in power there even though it didn't stutter (and it was going to about 14.3-14.7:1 there), so we felt it was safer to have a slight stutter occasionally, but have staging done by the time the engine wakes up at the higher RPM. Looking back we could probably have left the staging settings alone without the stutter, and just richened up the parts of the table where it staged to compensate for the slight lean spot.

So you're going with the S5 manifold then, that's cool. Are you going to leave the VDI on it? Going to re-enable your aux ports?

I got my copy, it's a good article. Tell Per good job.

Ken

 

Thanks Ken. I will let Per know.

I am going to go with the S5 intake and figure out how to operate the VDI and 5th & 6th ports. The ports shouldn't be a problem using an external pump and I am still reading up on how to use the VDI. Any suggestions on either?

 

When we had them running on tofuball's old car, we just used an external electric pump, and used spare outputs on the MS to drive the solenoids. With MS2, I'd use the IAC outputs to drive relays or transistors, which then drive the solenoids for aux and VDI. The pressure from the pump was used to operate both the aux ports and the VDI if I remember correctly.

We had a pressure operated switch to keep the pump from running all the time. Any time pressure dropped below 5psi, the pump would come on.

For aux ports we opened them at around 3800 RPM I believe, and we engaged VDI at 5500 RPM.

You can operate these by using IAC1 and IAC2 in the spare ports dialog in megatune/tunerstudio.

Ken

 

Damnit Ken. I forgot that you live in the DC area with your acronyms and vastly superior intellect.

So I would use the Idle Air Controls 1 & 2 to operate an electric pump to trigger both the Aux ports and the VDI. I would think that you would need two separate pumps to for each so they could each supply the correct pressure at the correct times.

I am also assuming I will need to change my map significantly with these changes. I have only scratched the surface am probably drowning.

 

Good stuff, I'm just watching the movie now.

Why didn't you use the primary staging reduction delay with more cycles before reduction? Its also possible your values for your injector cc are wrong too I guess.

I stage at 3500rpm and 80kpa in my 12AT mazda and have no problems. I think I use a 10 cycle no reduction, with gradual over the next 2 cycles. Something like that. Its stumble free here, I can't even tell when the secondaries come online.

 

hrmm, I might have to try those settings on my TII. I don't really stutter most of the time, but the it's definitely noticeable when you're looking at the AFRs. What do yours AFRs do?

but yeah, it was a good article. I enjoyed it. I even read that before the berzerkly article, and that's quite a complement.

 

Staging at low rpm can be problematic as the fuel may be squirted into the secondary runners while the secondary throttle plates are still closed. Some have found tps-based staging to work better with progressive throttle bodies like that used in stock 13b's.

 

In the article there was a picture of the MS2-v3.57 ems with a additional board installed in the case lid. What was the board? I looked on DIYAutoTune's web site for mod boards and can't find any thing that looks like it.

 

Just read up that article today actually! Very nice and good job

 

I couldn't give two ***** what my AFR's do. My car has survived countless power runs, and really many times my driving is worse than anyone would treat a hire car. I broke an axle and twisted the other not 3 days ago from enjoying myself.

If you want to be a girl about numbers maybe you should get a new hobby like playing Yahtzee?

Sorry but there are things that work, and there are things that look pretty. And if you want to base 'the quality of my tune' around what my wb02 sensor would be reading rather than how the car is driving and making HP well

 

??

Not exactly, we just had a single pressure switch and single air pump with the IAC pins controlling solenoids. Whenever the pressure dropped too low, the pressure switch would turn on the air pump. That pretty much meant that any time one of the solenoids turned on, the pump would run briefly to pressurize the lines again, and actuate whatever that solenoid actuated.

Yeah, your VE table would change considerably. I'd say another visit to the dyno might be in order.

Ken

 

Duty based staging at 70% was actually working and just going ever so slightly lean. You could see a power drop on the dyno, but you couldn't feel anything.

We switched for the reasons mentioned in my first post in this thread... In my experience duty-based works better than any of the other three options.

Ken

 

It's the board I designed for my plug-n-play units.

I think Jerry may start selling them.

Ken

 

OK:

1) Attacking other forum members will not be tolerated.
2) I want the AFRs to stay perfect because it means my code is working well. IF they're not perfect, it means I still have work to do on my code to MAKE it perfect.
3) Even if I wasn't writing the code and Myk (eage8) wasn't a good friend and coworker of mine, I wouldn't want the AFRs even going briefly lean on boost. I don't care if the car runs alright or not, I don't want to ever go lean while in boost.
4) When it *does* go lean, you can see a dip in the torque and horsepower, so even if you don't FEEL it, it is affecting the way the car runs.

Some people want everything to be perfect; "good enough" isn't perfect.

Ken

 

WE actually started with duty-based staging but with similar reduction settings. it worked fine except for going a little lean right where his power started to pick up, and affecting how quickly power picked up.

In any case, if we really needed to get rid of the lean spot we probably should've just richened up the VE a little in that area. In any case, you couldn't *feel* it on the butt dyno, but you could see it in the AFR and on the dyno plots.

Ken

 

So on ms3 over the past weekend I came up with a new staging method that appears to work much better than any of my previous ideas for staging. Looking at AFR with this new method you can't even really tell it staged.

The idea is that instead of MAP or RPM or duty or TPS or anything like that, you stage based on a table that is load on the y axis and rpm on the x axis. The contents of the table are numbers between 0 and 100% with 0 meaning don't stage at all, and 100% meaning fully staged (so equal pulse-widths on primaries and secondaries).

On my GXL (S4 NA) I set it up to stage at 4000 RPM and 50kPa in a box in the table, starting at 10% staged at low load or lower RPM, and climbing to 100% staged at high load/high RPM.

We also tested this on eage8's S5 TII with it set to stage just before boosting, starting at 30% and climbing to 100% as boost climbs.

With either car any stuttering is essentially gone and you can't tell by looking at the AFR that the secondary injectors came on.

I did have to retune my VE on the GXL (3-5 points anywhere it staged) but I think more accurately figuring out and setting the opening time on the injectors would've avoided this.

In any case I'll be backporting this staging method to ms2 this week so that people on here can try it. It'll go into 2.1.1.

Ken

 

thanks for youre work!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

 

+1

I haven't even installed my MS yet but since i put it together last year he has made countless upgrades and has been super helpful.

 

No problem. If I didn't like doing it I wouldn't!

I just wanted Greg (HadaGSL-SE) to know that I think I came up with a solution for the occasional stutter he was having on staging.

Ken

 

Thanks Ken. That studder doesn't bother me as much as the light throttle bucking. I was thinking of driving the car up to NOVA for Christmas this year (tight fit with the wife, dog and luggage) and would love to meet up.

If not I should be able to make the Megameet 2010. I am still looking to do the S5 intake and try to pick up a few more ponies. Or at least get the 5th & 6th ports working again.

 

Yeah, I think to help with that light-throttle bucking I'm going to need to see what's causing it.

Did you ever get the pulley issue worked out? On my own rx7, richening it up a bit in those areas and getting the timing a bit more advanced got rid of any light-throttle bucking. I read in a document that I believe mike_robert posted that overlap and other issues can cause some misfiring at low load and low rpm if running too lean.

Fixing the aux-ports might also help make that easier to tune in case it's just an air velocity problem or similar.

Ken

 

How does one do that (figure out the opening time)?

 

You can hook the injectors up to a scope and actuate the injectors with the scope hooked up. This will tell you how long it takes the coil inside the injector to saturate. I usually start with this value. This doesn't tell you how long it takes for the injector to actually physically open though (although with good clean injectors, it's very close).

You can also adjust the number of squirts, then at a mid-high load, changing the number of squirts shouldn't change the AFR. You can't really do this at idle because fuel flow isn't necessarily linear at low pulse-widths. This method has worked alright for me in the past. For the secondaries, I would just adjust it until the engine runs the same AFR with all 4 injectors going as it did with just 2.

In any case it doesn't matter much as the amount of retuning necessary was minimal, and the new method works so much better on the engines i've tried it on.

Ken