Any general discussion around the firmware, what is does, how it does it etc.
#61198
It runs! After fixing a leaky fuel filter line, properly running a ground from the coils to the Speeduino and cleaning the injectors; I cranked it for a few seconds, it started right up for the first time in 10 years and idled! This is running LS1 Coil Per Plug with the decoder in the pull request in Git Hub. I went into the day with 80% confidence about it starting and am glad to report I am impressed with how well it all worked together. Now I need to figure out what to do with the valve cover breather and separate vacuum line coming from valve cover. Then I will move on to tuning, which I have never done before.
#63091
Attached is MSQ. It is a baseline tune for fuel map, timing and such but has all the trigger setup, engine constants etc. It is the same tune I used to drive it around the block, the only change since was setting it to Wideband O2. I haven't actually installed the Spartan 2 02 I now have. I am looking forward to installing that and then checking timing with a timing light because I suspect it needs some timing table adjustments.
Separately, here is a side rant from a theoretical perspective because this will be the first engine I have tuned. Maybe this is more of a Tuner Studio commentary rather than Speeduino, but I am sure this community can address it:
  • Typically timing adjustments seem to be done by ear/seat of pants in small amounts
  • If you have a dyno then it is done based on the dyno results
  • Autotune only adjusts VE table
  • Is there a log/analytic or can someone create a way to analyze the acceleration of the vehicle to determine if it improves after a timing change?
    • I imagine from logs one could compare speed/acceleration cross referenced to the timing table (maybe try accounting for gear) and tell if there was an improvement
    • I know this would be effected my road condition/slope/wind/weight of load in vehicle but it seems better than just seat of pants tuning if one is inexperienced and doesn't have a dyno
Attachments
(87.48 KiB) Downloaded 325 times
#63101
RowdyDouglas wrote: Sat May 20, 2023 4:43 am Separately, here is a side rant from a theoretical perspective because this will be the first engine I have tuned. Maybe this is more of a Tuner Studio commentary rather than Speeduino, but I am sure this community can address it:
  • Typically timing adjustments seem to be done by ear/seat of pants in small amounts
  • If you have a dyno then it is done based on the dyno results
Tuning is always done with data. ECM data, dyno data, sound, smells, etc, but always data. So yes, ears and seat-of-the pants are useful to a limited extent, and mostly as verification of what the other data is telling you; but the data from your ECM and logs is the primary source for info to guide tuning. When processed through MegaLogViewer it can create a form of "dyno" data that is very useful to guide tuning, whether on the street, track or dyno.

Tuning is always about efficiency. Efficiency enables more of whatever you are after. Torque, horsepower, economy, emissions... whatever it is, it is best at peak efficiency. We use data that indicates efficiency in whatever mode we are tuning, and there are many indicators.

RowdyDouglas wrote: Sat May 20, 2023 4:43 am
  • Autotune only adjusts VE table
Correct. The parameters for other tuning is a bit too much for current versions of TS and MLV. Bummer, as it could be capable, but the biggest hurdle is training users to use it effectively.

RowdyDouglas wrote: Sat May 20, 2023 4:43 am
  • Is there a log/analytic or can someone create a way to analyze the acceleration of the vehicle to determine if it improves after a timing change?
Sure. There are many modes of operation (idle, accel, cruise, power, decel, etc), and acceleration is just one. Look for what data would change or indicate change, in that mode and under those conditions. Watch that data. Examples could be rpm/s, or time vs rpm, or speed over distance, etc.

RowdyDouglas wrote: Sat May 20, 2023 4:43 am
    • I imagine from logs one could compare speed/acceleration cross referenced to the timing table (maybe try accounting for gear) and tell if there was an improvement
That's a good one, especially if simplified. Consider various ways to measure what acceleration is. Use data to define it. Plot that data. Run multiple identical tests comparing each to changes you make to timing. Use the timing that indicates best acceleration. This is the process of "test 'n tune" using comparatives, and is a form of diagnostic tuning. The basis of diagnostic tuning is to compare results, from bad to best and back to bad. Now best is easy to identify as the best setting.
Find BEST Tuning-Diagnostic bracketing.png
Find BEST Tuning-Diagnostic bracketing.png (16.1 KiB) Viewed 4183 times
RowdyDouglas wrote: Sat May 20, 2023 4:43 am
    • I know this would be effected my road condition/slope/wind/weight of load in vehicle but it seems better than just seat of pants tuning if one is inexperienced and doesn't have a dyno
Even professionals find the butt-dyno to often give poor or even incorrect results. :lol: Other forms of sensory data can still be useful, such as adjusting idle timing and hearing the rpm rise, indicating increased torque. However, this is always verified with data. Use more than one source of data whenever possible, i.e., knock sensor, amplified ears, EGT, reading spark plugs, etc, or some combination of them. The skill isn't so much reading them as identifying and using the best or most-appropriate data sources — what indicators (data or data sets) will tell you what you want to know?
#63268
Thanks for the base tune to check out. I will probably be using 4 honda K series coils as that is what I have laying around....

Did you verify with a timing light that what the timing is set to, is what is being output to the coils? just curious. when I try this on my jeep I plan to verify this just to make sure there arent any bugs....

Timing.....

There are optimal values that make the most torque at a give load/rpm up to the point where you must reduce ignition advance due to being knock limited. Then there are safe values that you can plunk into the timing table and have it be good enough, not knock and run well.... you may lose a couple of HP but it will be safe.

ive seen people use this software to analyze a high sample rate to see if the rate of something like a vehicle speed sensor or RPM increases at a faster or slower rate indicating more or less power. https://barnhill.bitbucket.io/ you will need to try this locked into a certain gear and always on the same road on a day without significant wind so the data is valid. A jeep 2.5 doesnt have alot of power so wheelspin shouldnt skew the data :)

https://uprev.com/documentation/Knock%2 ... Device.pdf

this setup works to listen to knock. I use a variation of it to listen to the knock sensor on my honda. I used a Fiio A1 headphone amp rather than the E5 in the PDF. the A1 works well. I can sit there with the car off and move the gear shifter and hear the gears engaging. you can hear everything in the engine... especially faint knock before it becomes big knock.

Brian
#65314
@Bradleyiii, to recap one of my pervious posts: it ran and I drove it around the block. Realizing it needed tuning I installed a 14point7 Spartan 2 wideband O2 and updated the project file (around the same time I installed a cable actuated axle engagement device which included some wiring).
NOW, the Speeduino is only reading about 4v but my DMM reads 12v. I tried disconnecting the O2 sensor but I need to unwire the controller and look for difference
OR the problem could also be something with the axle locking wiring like shorting a shared ground or something.
Sidenote, I also took off the stock (pulley mounted?) fan and shroud to replace with an electric fan but haven't bought nor touched wiring for it. #nowimcommitted
  • 1
  • 6
  • 7
  • 8
  • 9
  • 10

Hi, I am trying to assign Signed values to the x-a[…]

Vr Conditioner Noise when cranking

New version 202305 don't fix the issue. Now after[…]

blitzbox

I've finally figured out why MAP and Lambda weren'[…]

Hello, I bring news!! Let me tell you that after […]

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