Second firmware live stream happening this weekend: viewtopic.php?f=13&t=3992
Help with building your Speeduino, installing it, getting it to run etc.
By jonbill
If I understand right, you've effectively captured an AFR table from your D-Jetronic, where the D-Jetronic runs nicely.
and you're going to tune Speedy to that same AFR table to hopefully make it run as nicely as the D-Jetronic.
is that right?
I think it should go a long way, but there are other variables I think that could still mean maybe it won't be as 'right' as D-Jetronic. I'm getting out of my depth now :)
By Torslund
That is what I do for a starting point, make it run as Djetronic and then I can slowly tune in the speedy.
By Black Knight
Torslund wrote:
Fri May 22, 2020 12:04 pm
That is what I do for a starting point, make it run as Djetronic and then I can slowly tune in the speedy.
If it were me, I would get AFR data and ign timing data from the d jet and then a reasonably good VE table from speedy or another tuner to get it started. Then let auto tune and hand tune do the good job they can.

The above is dependant on good sensor input and calibration like a standard start from the beginning build and tune.

I did similar with mine. I had two known good running carbs (logged the afr target) good data, then loaded the afr info on the arf target table.
I had a good tuned dizzy and logged the timing and and put that on the ign table.

Found a fairly good VE table and let TS do its job.

The only thing I did wrong was not use all sensors and calibrate them so the sensors were changing afr and timing.

Got that fixed with the help from good people here and started learning how to set up base tune.
Then there is the technique of tuning to make it all work right.

I am not sure here but the d jet may use the data differently than the speedy so for an example your d jet idles at 12 afr and speedy may idle at 13.8 to 14.7.

I hope someone smarter than me could chime in on that one.

I am just trying to get my head around where you are with this and maybe find something to help you out.

Black Knight
By Black Knight
Torslund wrote:
Fri May 22, 2020 9:11 pm
One ting that's iknown, V12 don't like running lean. Goes warm and valve seat droops down.

You're not giving me much to go on. If you want some help you're going to have to share more detail.

Black Knight
By Torslund
The sensors, IAT is the one that comes with the car, I have the value and have measure them also, same with the CTS. MAP is the one that comes with the speedy board. WB sensor AEM. TPS is from a XJ40. All sensor are calibrated in TS. If they don't do what they should, how to know?
My thought is, give engine what it was designed for. Make it run as it do on the Djet. And from that, use the modern electronic for fine tuning. The Djet VE map has not so many cells, Speedy has more cells. So it should be more accurate.
I have the WB and with that I can read mixture.
Not sure what the auto tune does, have try it and it started to make changes till engine did not run anymore :)
User avatar
Torslund wrote:
Fri May 22, 2020 11:40 am
Ok, but if I tune VE so the AFR reading is what I want. Does that work?
Respectfully - no. You want to tune the engine to best operation at one point (rpm/MAP), then see what the AFR is there. Enter that AFR into the AFR Table at that rpm/MAP location. Do not tune to an AFR, unless it is the one you have tested to know is the correct AFR for that engine, with that hardware, that fuel, etc. Testing for best operation, and from that knowing the best timing, fuel, and target AFR/Lambda at any point is the goal of tuning. Give the engine what it wants, not what you want or somebody said on the internet, or what the old ECM did when designed for different fuel, parts, and old emissions requirements.

If you do this, beginning at warm idle and moving to other parts of the tables, you will have an engine that runs better than D-Jet. Just get it running well enough it doesn't quit, turn off O2 corrections, and begin testing. At warm idle, adjust your fuel to show lowest average MAP at that rpm. Note the AFR. Move to the next table area and repeat. Once you have found the AFRs it really wants under different conditions, then let it auto-tune between those points with MLV, approving each change. Do NOT use live-tune until it is very close to finished, as it can make your tune worse. This is much quicker than trying to make it run like a D-Jet first. ;) Good luck!

By Black Knight
Good input David.

I agree that you need to do hand tune to get a base line with tuning. I agree he will get a better tune with Speedy.

I think he could use his after market wide band to get AFR target data and he could log his his basic ign timing. It will not be 100% correct but will get closer that guessing. Some where to start.

I am still running my timing off of the data I got from a well tune dizzy I set up. Very little changes needed for Speedy.
Just loaded the data onto ign table. Found some small changes with the great fine tuning with Speedy. Very nice.

Did the same with fuel. There was a lot of changes to fuel because speedy is soooo much more capable than a carb.

I also do think the D-jet is using the data differently than speedy will. So the data should only be a crude base line maybe.

I like to find target afr and timing for the no load area by hand tuning say between idle and 4500 rpm to give an idea of what is needed and have a place to get back to that is kinda running ok.
Some place it won't die. Then like David said build up and out from there.

I think the biggest problem now is there is no posted MSQ.

The tables and configuration could be off and we can't see that so it is hard to help out.

That is my next input.

Post the MSQ so we can look at the tables and how you have set up your tune so we try to find the faults.

There are good people here that know more than me on this but I will do what I can because the project looks interesting and I learn a lot from helping others.

Thanks for giving more info. It helps.

Black Knight
By Black Knight
PSIG wrote:
Sat May 23, 2020 8:33 pm
Torslund wrote:
Fri May 22, 2020 11:40 am
Ok, but if I tune VE so the AFR reading is what I want. Does that work?
Do NOT use live-tune until it is very close to finished, as it can make your tune worse. This is much quicker than trying to make it run like a D-Jet first. ;) Good luck!


Also with great respect, I did agree with this until very recently. I have been working on something that might change this. Your approach was how I did mine. By hand first. I also think it is the correct advice for now particularly for beginners and also as I have not perfected it yet. Very close now. It will happen.

I have spent 4 months working on what I call the base tune. If all configs and calibration are set right, I have had live tune do a fair to good tune from the start, after idle and some low load is done by hand. TS and Speedy are amazing if you give them good data. Way better than I thought in the beginning. You do need reasonably good target afrs and ign tables for this to work. The good basic tables Speedy supplies with the firmware has worked for me. Yes you will be making changes to these tables for you own tune needs later. This is normal.

I am days away from sharing this now. It will need a video to show how I did this and I would very much like your (David) input on this before I post it on this forum.

I hope this is taken as a possible new and good thing for us all.

I now know TS and Speedy can be amazing.

If there is an interest in this I should take it to its own thread and not interrupt this one.

Black Knight
User avatar
Black Knight wrote:
Sun May 24, 2020 12:56 am
… You do need reasonably good target afrs and ign tables for this to work.
My apologies to @Torslund for veering off in your thread. :oops: Perhaps I didn't word it well. TS and MLV are great, can work really well, and that is not my point. I am saying that it of-course needs to run, but then skip right past "better" and go directly for best. This means not auto-tuning the tables to get it running generally better. Just get it running and warmed (the purpose of the Base Tune), so you can then set idle and test for the correct target Lambdas (actual tuning). You can't begin to adjust priming, cranking, ASE or WUE before that anyway, so just get it done. Once you have the correct Lambdas for your project (not just "reasonably good"), then let it auto-adjust for you to those best targets that you found by testing.

To recap; tuning is finding the best operation (efficiency, HP, TQ, mileage, etc) for your specific project and fuel. Not "close" or some generic AFR target (14.7 here and 12.5 there) and timing (10° idle and 24° WOT or something), but best, and found by testing. This is the part you'd do later anyway, so just do it right up-front and skip the generic target stuff. Does it make more sense if I say it that way?

Note that adding the step of tuning to general targets is good practice for new tuners, especially if they are not comfortable with the programs, or have not learned the methods or developed the skills necessary to find best timing and Lambdas right away. Confidence is a good thing to develop. In that case, then sure, add the extra step for that practice.
  • 1
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32

From digging around, I think I found my issue... P[…]

I fixed the timing problem, it has no jitter now. […]

New 202005 firmware

is it possible there's something particular in you[…]

Yes although a simple trigger for fuel only will s[…]

Still can't find what you're looking for?