Adaptronic Boost controller setup.
 

Need info on how to setup Mac boost controller valve? This is for single turbo application, OmP and twin turbo boost controller have being remove from harness, but have 2 spare wires available in my ReWire harness that I can use.

Basically, disconnect the wires going to the factory RX7 boost control solenoid, and connect in the MAC valve instead.

Then plumb it normally and set it up in the ECU as usual.

Here's an article / video on how to set up the boost control, boost by gear, etc.

Setting Up Open Loop Boost Control | Adaptronic - Take Control

Thanks!

Andy,

As I stated on first post I have a single turbo harness, all of the wires for the twin turbo control have being removed, i have being using a Greddy boost controller to control boost. Now since im going to have the Adaptronic control my boost, I will have to rewire the ECU connector to add the wires to control the boost solenoid valve.

Looking at the ECU wiring pinout I see that pin

4R = as Selonoid Valve Turbo Control
4S = as Selonoid Valve Charge Relief
4T = as Selonoid Valve Charge Control
4U = as Selonoid Valve Wastegate Control

Also Pin 4A-B are constant ecu ground output.

Which of the Pins will I have to recconect to my harness?

Thanks
Luis

Hi Luis,

It's the wastegate control output you need.

So connect one wire from the MAC valve to pin 4U, and the other wire from the MAC valve to an ignition switched +12V (eg grab it from another valve like the idle valve etc).

Thanks!
Andy

Bingo... doing this today =-)

I wanted to add here,

I asked over at the Adaptronic site how I could setup a two stage boost function. Andy recommended to simply put an inline switch for the MAC so when switched off, the car would see spring pressure.

Yep, that's the easy way if your low spring pressure is suitable. If it's not then you can hook up a switch into a digital input on ECU and set it to "boost 1" and then change the maximum MAP setpoint and duty cycle for that boost setting. You can do the same for "boost 2", and putting them both together gives you boost 3. In my car I have both that and ability to switch off the boost control valve so I have 5 possible boost settings.

Andy, do you recommend opening the ECU signal, or the 12v side of the MAC? I assume it doesnt matter....

Yes, it doesn't matter; whichever is easier. I would assume you'd be getting 12V from within the engine bay for the valve so it's probably easier to intercept the negative line to the boost valve.

Hey guys having a little trouble with this one. Hooked everything up yesterday using the mac valve. One wire to 12v switched and other wire to pin 4U on the ecu. In the wastegate settings max duty cycle is 100 and duty cycle is 40 percent all the way down but i'm not sure where this should be set at. Boost is at 15 psi. First test pull it sounded like the wastegate was very slow to open and boost kept rising and spiked all the way to 19 and probably would have kept going if I wouldn't have let out. As far as plumbing goes I have it setup the same way as I did on my profec b spec II, off the compressor housing a line goes into port 1 then out port 2 then goes to the side port of the wastegates and I have a line from the UIM connect to the top ports. Any help would be appreciated!

depends what your wastegate spring pressure is. mine is around 15.8 and I dont need very much duty cycle at all to reach 17psi.

Start the WG settings at ZERO, and confirm you are not over boosting. I assume you didn't over boost before? A lot of turbo manifold designs tend to boost creep FYI. Always set your soft and hard cut boost settings to a few psi above your target while setting up the boost controller to keep from popping your engine. Soft cut is usually set 1 psi below the hard cut.

I'm having a bit of trouble with the wastegate control on my Select 440d.

I've just got it up and running on a 12a turbo with a t04z and Tial MVR. I'm using the MAC boost solenoid and it tests fine (clicks when inverted). However on the wastegate screen, the desired map is all grayed out at 15psi. Now, I initially only want to run low boost, say 6-8psi and I've got the boost cut set at 10psi. I can tell the wastegate isn't operating because there's no noise from the screamer pipe.

I've been messing about for hours trying to work out which setting I've got wrong but I can't for the life of me sort it out. I've attached my ecu file so you can tell me how silly I am for making such obvious mistakes haha. Any other pointers would be appreciated.

Update; It appears to be an issue with Wari. I was using V13 and when I was checking some base maps on my desktop which has V12, I noticed that even my current tune appeared to function correctly on the wastegate screen.

I have just revered back to V12 Wari and it seems to be working. About to go and do some testing.

Pretty disappointing stuff, I expected more from this ECU. Also, the idle valve doesn't work. I'm not sure why but it's definitely software/firmware related.

The valve tests fine. I can increase the minimum value under the idle tab and the valve responds accordingly. However, it doesn't matter how I try and manipulate the open loop base idle value (based on coolant temperature) the idle valve will not respond. Extremely frustrating.

I realise the above ecu settings file has a few wacky attributes, but it is out of mere desperation to try and get the thing to cooperate. Once again, any constructive input anybody has, I'm all ears.

Also, I cannot improve throttle tip in. It goes extreme lean every damn time. 22:1 AFR. Also, every time you back off, the car stalls. I'll upload the current settings shortly. It's going absolutely fantastic on boost, but the before mentioned things make me want to throw it in the bin.

How is your vac ref plumbed?

Are you using map prediction or the older accel enrichment?

Does the throttle % read smoothly in a log? TPS calibrated correctly?

Do you have MAP filtering enabled?

Can you post your current file and setup details?

Are yo using VE or ms maps?

Injector dead times accurate?

Plenty of people have them running fine, better idle running and transitions than most comparatively priced ecus, there must be something not set right for you to be having dramas. I recon we can sort it without having to reflash firmware or anything like that, but worst case, reload firmware.

Vac reference for the ECU? It's just plumbed up to a vac nipple on the back of the series 4 lower inlet manifold next to the rear rotor secondary injector. In hindsight, perhaps a bad spot?

For accel enrich I'm using MAP prediction as thats what all the FD/FC basemaps seems to have and everything I've read so far suggested to use it. At this point I've mainly tried adjusting the asynch gain and double asynch gain to no effect. When I crack the throttle open it goes extreme lean every time. It may also be worth mentioning that my primary injectors are mounted on the lower inlet manifold in line with the secondaries. So they're in a less desirable location that a normal injected rotary, but I don't see it being an issue as they're no further away than an injection perfection throttlebody or similar.

Yes, the TPS is nice and smooth in the logs. Using megalogviewer for visual representation. TPS is calibrated correctly.

Map filtering is set to > 70kpa

Current file is attached. The setup of the engine is a 12a extend port, SDR t04 turbo manifold (shitty wastegate pipe design but holds boost at 6psi just fine directly off the wastegate spring) T04z turbo, s4 lower inlet manifold with s6 upper, 4x ID1000's, chinese intercooler, innovate mtx-l, select 440d... let me know if I've left out critical components.

Using VE maps

Injector dead times I haven't touched as the ECU has that pre-done for ID1000's, at least I thought it did.

I know plenty of people are running them fine, that's why I decided to go for something like this instead of megasquirt again. I didn't feel like having terrible headaches, but here we are, terrible headaches with basic stuff that should be right out of the box. After seeing what happened with the bleedingly obvious wastegate values being grayed out using the latest V13 download, I'm left wondering what else is wrong with the software/firmware on this thing. Well for one thing, I'm pretty sure the idle control part of the software is fucked. What else?

Thanks a bunch for your input mate, I'll be interested to hear what your further thoughts are.

This picture illustrates the idle valve's unwillingness to cooperate, or more accurately, the computer's unwillingness. You can see the botched up table calling for 90 idle effort, but the gauge window clearly shows only 60.
https://i6.photobucket.com/albums/y2...dleproblem.jpg

EDIT: Just flashed V12 firmware and whaddya know, the idle valve works as it should. So in conclusion, V13 is a steaming pile.

I'm on v13, and got my idle pretty well in open loop, still working on it before i move to closed loop.

Prior to me getting to work well, I had "ExtIn 1/64 Elec Load (NB MX5/Miata" checked on accident, which sounds to me like what you are going through. Coolant base idle and all load inputs were not functioning or making the IAC move at all, it was just stuck at the minimum value.

Thanks for your input mate. I can't seem to find the checkbox you're referring to. Which tab is it under?

I think this is in the v13 software only. I only started playing with idle control in v13.

It is on the idle screen, it is adjacent to the "extra effort for elec load 1" or under the coolant based idle box in your screenshot.

Edit: Just looked at your map in v13 software, and that box is indeed checked.

Your predicted map table looks the same as mine. Are you seeing the lean spot via logs, or while driving? Here is one of my logs from earlier today. As you can see it is about a half second to full second before it returns to normal AFRs. This was from a "Decel fuel cut" or off throttle cruise to back on throttle. There is no noticeable affect in driving, but I can see where you are alarmed.
372.951 1292 -5.92 0 86 190 196 - -
373.31 1333 -3.88 0 86 190 196 22.3 13.1
373.404 1321 -4.25 0 86 190 196 22.3 13.1
373.731 1378 -4.77 0 86 190 196 15.3 13.1
373.887 1404 -5.03 0 86 190 196 14.9 13.2
374.152 1402 -5.14 0 86 190 196 13.7 13.2

I've played with my predicted map table to get it smooth, and the logs still show the above. I want to attribute it to lag of the AFR sensor, but as you can also see, it is about a full second and it is back to normal afrs.
My injector times once I get back on throttle also make sense, so I'm not too worried with that shows on the log.

BTW, also looking at your map, I believe you set your injectors wrong.

It has been stated by Andy in a video and probably some literature he wrote, but for staging to work, you need to set your primary injector type to "dead time/flow" under the tuning mode tab, then under the corrections tab you need to set your dead time corrections for "Primary" and "Secondary" should be straight forward here.

No air temp compensation for fuel? May want to throw in an ideal gas law line?



Is there a vac ref just after the throttle body or on the plenum going spare? It has been a while since I looked at one, probably better than off an individual runner.

I have revised some stuff in the file with notes attached.



Closed loop boost control :lol: >> If you put integer values in the proportional gain (can keep zeros in the integral and diff if you wish) you can then edit the closed loop target boost values in wastegate control. Basically it can't work closed loop without at least a proportional PID parameter. There are several threads on this (regarding people observed good PID values on different setups) on the adaptronic forum.

It may pay to use a finer mesh for mapping, not so much for revs but having a denser map around idle speed means you can run a "timing" well so that idle speed settles and if it drops the additional torque of a bit more timing just below target speed catches it. Is your timing true crank degrees or offset to the ATDC mark, ported motors seem to be happy around 7 deg true at idle? I did this on the Haltech in OLD MAN's FC (he was asking for help on AR) which had IAC disconnected and I cleaned the idle fuelling up a bit (some numpty had it on a "closed throttle map"). Got it starting and idling happily at OEM revs with no IAC at all :lol: I will note his EP 13b idled happiest around 12.7 AFR on the gauge. Even if you go down to 250rpm increments across the whole map that only takes you to 7500 IIRC, the ecu carries the last table values to higher rpm so you could still run to an 8000+ rpm limit, would obviously go a bit rich there. (I'm surprised you aren't revving it out more, even with a 500rpm optimistic tach I generally run my 12a off the end of the dial and an running out to 8350 with the PFC in the FD which is near stock)


As above, does the engine actually stumble on throttle transient or is the wideband just slow to catch up? It should still not read lean at all (unless throttle off overrun is active) if the map prediction/asynchronous gain is perfect.

Just noticed the predicted map at low rpm low throttle openings is arse about to what it should be, it will go closer to atmospheric at low throttle openings at low revs than it will at high, table needs tweaking. You can do some logging playing with throttle at various rpm to find the pressures it will sit at when you crack it various amounts to improve the table.

Thanks for that slides. The base maps I had must have been done in V12 software and the stupid Miata setting was absent with it checked as default when the V13 software was installed. Frustrating, but glad to be on the other side of it.

I changed the vac reference point to one near the throttlebody, made no difference to anything.

Closed loop boost control is working really well, holding 11psi nicely. As for the rev limit, I've started with it very conservative as I want it to live and just last week it ate it's torrington bearings due to an incompetent engine builder, so I don't trust it yet. I literally did the whole turbo conversion in one day, so just building trust in the system.

My timing is true crank degrees.

The engine stumbles on throttle transient, badly. The car is basically undrivable. I spent 3 or 4 hours trying to get the car to respond to throttle blips with a cluey mate of mine but it absolutely refuses to cooperate. Tried both map prediction tables and the ol' style. No luck. I had a look at a bunch of people's tunes, tried their settings, tried everything from high to low values, different transition times, asynch gains etc. Car drives like complete garbage.

Also it stalls every time without fail when you come to a stop and I can't for the life of me work out why. I'll get a log up at some point but I can't see any reason for it. The AFR's look fine, timing looks ok. It just falls over. Really ready to burn the car to the ground haha.

For the mean time, here is a video of the rx7's spectacular throttle response. Couldn't really improve beyond this, no hope of getting it to rev cleanly. So far as I know, there's no Adaptronic gurus around here either.

Might be worth checking TPS calibration again or your threshold limits?

Have you tried it with the TPS sensitivity at different levels?

Maybe try some predicted map values a bit more like attached if you haven't already?

Tried toggling the tick boxes for open loop transient etc?

Have you had someone stand behind the car to see if you get any soot when you blip it? Or there is nothing?

Seems really strange.

Could you post a log of you free revving it like that?

Might be worth playing with the closed loop/dashpot idle checkboxes to see if there is any influence there for the stalling?

Would be really interesting to see what actual inj ms open times you are getting on transient vs steady state on load at those throttle/rpm points on the main fuel table.

Thanks again for your input. I have quadruple checked TPS calibrations, so I'm sure that's all good. Threshold limits are 1% and 96%, if that's the setting you're referring to.

I haven't tried different TPS sensitivity levels as I can't locate them.

I have tried predicted map values like those you have posted. I just did a log with those values for you, but as per the settings I have just uploaded all the check boxes are ticked as it just won't respond otherwise and it's still horrible like that.

Yep, tried toggling the tick boxes and values for hours last night to no avail.

I'm in the process of having someone stand behind the car, should be able to do it this afternoon.

I haven't looked at playing with the dashpot settings yet, but it's something I may need to look at. I've uploaded 2 logs, one sitting stationary whilst blipping the throttle, the other going for a quick drive where it of course tries to stall, then eventually does. There's sweet FA I can do to stop it stalling because the acceleration enrichment just won't work. So frustrating.
Anyway, hopefully the logs reveal something that can help me. Also I know there is something up with the water temp sensor, but that isn't causing my problem.

On a side note, this car is my daily haha, I'm desperate to get it to behave.

Apart from the coolant temp dipping and associated rev limit occasionally it all looks to be reading ok?

Not a lot of throttle pump effort showing however with TPS changes?

Might be best to email all that to the adaptronic tech address to see what they think, I'm pretty sure Andy and one of the other guys have a few different rotary files they can compare to.

Series 8 basemap settings for comparison, maybe need to wind down the gain value a lot?

I disabled the coolant based enrichments and WT idle up maps for now until I fix the sensor issue. Which is fine when the car is hot, and that's all I care about right now.

I did notice there wasn't a lot of throttle pump effort showing, of all the changes we made, we were absolutely baffled by how little it effected the damn thing.

However on the way home today my brand new air filter decided to fall apart and say g'day to my compressor wheel. As you can imagine, it's not pretty. The turbo is still functional but I know I should get it rebuilt :( So that's a whopping 4 days I got out of my turbo 1st gen haha.

I do have that series 8 base map on my laptop which I used as a reference, among others done by Turblown and some other dude tuning an FD.
I was focused on FD basemaps given I'm using the FD manifold and throttlebody.

You're right, I may have to get on to adaptronic and see what they think. I wanted to try here first as I wanted input from rotary minded people (which I got, thankyou) rather than getting generic information.
I know when I tuned my 13b bp on the megasquirt, getting the TPS type tune with accel enrich right was an absolute bitch and there was no real information to draw from.
At least Andy knows FD motors! haha.

Thanks again for your assistance, I'll try some more stuff tomorrow. I'm sure my neighbours hate me by now, especially with the throttle blips.

Michael: thanks for helping out!

Hi Evil, I have 3 guys fulltime whose main job is tech support - out of them Mark has tuned a few rotaries, plus I have obviously and I have the 13B in my car.

On the idle control, it can't do closed loop idle properly because you have the target idle set to 1000 and the minimum duty at 60%. At 60% it's idling at 1200 RPM, so it has nowhere to go. By the lack of vacuum I'm assuming it's a bit of a portey engine so I understand if it needs to idle at 1200, but we should put the target up to 1200 and bring down the minimum so the ECU can go into closed loop. I'm assuming that the clutch and neutral switches are hooked up because those inputs are configured in the ECU file?

Then I would enable some derivative on the closed loop idle, eg set the D gain to 1 or 2 or 3. I probably wouldn't go higher than 5. But with the D set up, when the ECU sees the RPM dropping, it will open the idle valve a bit to compensate so that should solve your undershoot problem on return to idle. You can tell that's working because on the F11 window, where it has the "flags" at the top right, we have "IDLE" and "FD" - "FD" will turn on when the throttle is released and the ECU sees the RPM dropping.

For the transient throttle, can you tell me about the setup? Is it an injected engine, or a carby engine to which injectors were added? Where are the injectors mounted relative to the intake ports and which injectors are you using? From my own experience the factory placement is pretty good for transients but I set up a 20B once where the injectors were ages from the intake ports "to get good atomisation" - maybe it did that but it made the transient throttle a pain to set up.

One thing to check is try it with the opposite "injector offset" on the triggering tab. It's zero in your file at the moment, so try it as 1 and see if it gets better or worse. I see there are some unusual numbers in the triggering settings so it looks like the sensor isn't in its standard location so that could be affecting the injector phasing.

Otherwise thank you for posting the log file and ECU file, it makes it very clear to me what's happening. I can see that the ECU is making async pulses at the right time, they just aren't big enough to overcome the lean effect. I can also see that you've maxxed out the settings related to async gain. If it's useful I can make a custom FW version with say another double the gain or 4 times the gain that you can try, before we make it a standard software option.

Hi Andy,

I suspect I may need a similar thing if I ever get round to fitting the injection gear and turbo to my 12a :blush: . It will be throttle body injection and everyone I have spoken to with decent post injection runner volume before the ports with a range of ecus has reported spending a lot of time messing around with transients to get them to work, I believe it is a combination of both injector timing, but more importantly on a transition into load, maintaining a consistent runner wall wetting (as they do get soaked in fuel) with wildly changing airflow/pressure so that the condensation/evaporation off the runner walls remains in equilibrium across the transient.

If I get to the point where I want aux outs on the FD I'll throw a select in that one day too.

Thanks Andy and Michael.



I don't have a neutral switch or clutch switch, those settings have carried over from the basemap I used. Sorry for the confusion there. Will open loop idle work without these present? It was my understanding that it may not. I also have no speed sensors. It's a nice and primitive 1982 rx7.

I'm yet to fully comprehend how closed loop idle works on this ECU as I've never used it before on any motor, sorry for my naivety here.

As for the setup, it is a 12a extend port (run of the mill high comp carby fed rx7 12a) with an FC lower manifold and FD upper. It lacks the standard primary injector location and the primary injectors are located in line with the secondary injectors. The injectors are genuine ID1000's. I had my concerns about the primary injectors distance from the inlet ports and have wondered whether messing with the injector timing would help here, however that's a bit beyond me at this point in time.
Attachment 629829
Attachment 629830

I will try the injector offset and report back. My triggering settings probably look weird because instead of moving the CAS to set the timing, I used the trigger offset to get the timing set correctly.

I'll do some more testing tonight and come back with a new .ecu file and a log. That extra gain firmware sounds like it might be worth a shot.

Oh, I've also managed to get the car driving a lot better by enriching the part of the map around idle, the problem is the car drives nicely but it wants to idle at 11:1AFR. I'm guessing this is where closed loop idle would be extremely beneficial? Or am I just plain doing it wrong?

Thanks again.

Please find attached .ecu file and log from tonight.

Suggestions?

you must have a lean spot

Hi Evil, can I ask you to send me an email please so I have your address?

andyw@adaptronic.com.au

It's a lot easier for me to help in that way because it's how I answer all other enquiries that way. It's also easier in terms of looking through the information in previous posts to find out what is direct information and what is opinions! Plus I can answer faster because it will just be there in my inbox.

To answer your direct questions though:

1) The ECU will work fine in open loop idle, but I was hoping to get the idle control to help with the revs dropping on return to idle. What we can do is set up the ECU to just go into closed loop for the RPM drop when it detects the RPM dropping quickly; we need to set that so it doesn't see it when you're driving because that will be very uncomfortable as a driver.

2) Yes, it's a wall wetting problem for sure with the injectors being so far away from the ports. I got around this on the 20B with some custom configuration (on the 1280 you can do that) and it looks like we'll need some magic on the 440 because I don't think the current system will squirt in enough fuel to work with this manifold / injector placement. We could try the original method (not MAP prediction) but I'd like to try with a lot more async first.

If it's OK I'd like you to email me so we can send files back and forth and I can help more quickly. I'll make up a new firmware version for you that you can test.

Unfortunately due to the nature of what we do, everyone's cars are different so we need to make it flexible enough to cover the majority of cases, but still easy enough to use. There is a tradeoff there and until we encounter a case where it doesn't work well enough there's no benefit to making the system more complex to deal with it, if that makes sense.

Thank you, I'll wait for your email.

Thanks Andy, emailed.

To Andy or any of the adaptronic guys reading.

So I was thinking about this yesterday, as a cheap and nasty "fix" in the current firmware/software, to get enough async on throttle transient could you just use very large predicted map values in the TPS table like 3 bar absolute to get a bigger hit or would the way the transient it transitioned back to the actual map reading mean it doesn't work? :scratch:

Hi Slides,

Adding artificially high numbers in the predicted MAP table is what I normally do when an engine is tricky to tune on transient throttle. But this is more of a band-aid solution, because if we open the throttle gently then MAP prediction may (or may not) be triggered which would cause the ECU to overestimate the MAP (due to the artificially high numbers in the predicted MAP table).

Thanks.

Cool, as I expected.