[PSIG]

… so I didn't think injector timing was all that critical.

It's not. That said, the injection end-point is adjustable, as the engine will respond better at certain angles for certain speeds and flows. Not huge, but it can be a bit better, sequential or not. A bit here and a bit there, along with a dozen other bits, and soon it's beyond good and approaching awesome.

Which bits you use depend on the setup and characteristics, and it turns into a big game to determine what could make the most benefit and work the list down. Most users don't tune well enough to notice, but some do, and so we have settings from injection end-point to multiple squirts to reduce lag or increase vaporization or improve fuel distribution, to individual cylinder fuel trims if that gets you what you need. Pick your team of bits and do your thing!

[Cgutierrez22]

Okay, so I found the connection issue was with the Speeduino Master code I downloaded from Github. I loaded the code unchanged and could not connect in tuner studio. I downloaded the 202207-fixes and loaded that and had no issues connection to Tuner Studio. The code had changed though, so I had to change all the Bit Set and Bit Clear lines to match this version. That is the good news, the bad news is with the TFI signature it didn't work right at all when first testing. I was getting crazy spikes all over the place and nothing was helping with that. I then switch to basic distributor and had issues with that as well, so I think the arbitrary waveform I created has issues. With a Squarewave signal, basic distributor was working like normal, so I switched back to TFI. In TFI, I was getting twice as many output pulses per input so I am going to work on adding a filter so only rising edges are triggered. I am also going to set the ignition end tooth part up in the code to see if that is the issue. For now, I am just going to use a TFI distributor as my trigger, until I can figure out what went wrong in my sig gen.

Chris

[Cgutierrez22]

I have the code working now and I used what Jama had thought would work. I measured the tooth width, and compared it to the previous gap, as well as comparing the falling edge gaps to the rising edge gaps. When you have the signature tooth, the falling edges are shorter than the previous rising edges. However, with a rectangle wave you could get it to sync by just being slightly off from 50% duty cycle, so I added a second check for sync that makes sure the gap after tooth 1 is 1.5 times as large as the tooth, and that this gap is larger than the last gap as well. So in wasted spark (because I don't have 8 channels) channel 1 always syncs to the signature tooth and of course it fires on tooth 5 as well. I also wrote the code so it should work for 4 and 6 cylinder engines with the TFI as well. I have only just started testing this, but so far it is looking good.

Wasted spark


Single channel


Chris

[Cgutierrez22]

Tuffin67, I have created a working TFI trigger. I have only tested it on the bench with a distributor, but once it warms up some I will be using it on the 5.8 in my F250. I added code for the 4 and 6 cylinder TFI as well, but those need more testing.

Chris

[5.0thunder]

Thanks jonbill, that is how I read it. My stock 351 only had 2 injector channels and ran fine with batch fire squirting 4 injectors at a time, so I didn't think injector timing was all that critical. The reason I want to get sequential with the TFI is I would like to run wasted spark to get more dwell and better spark at higher RPM. If I had a 302 I could just buy an Explorer damper with the 36-1 tooth ring on it, and cut all but the signature tooth on the shutter wheel and get full sequential, but the 351 is 28oz imbalance instead of 50oz of the 302.

Chris

In my previous research I found that you could probably also just cut the short tooth off the shutter wheel and run an 8-1 missing tooth cam and get sequential that way. It's similar to Signature Pip looking for a short tooth but it's looking for a missing tooth instead I suppose. Glad y'all are working on this, I have 6 fox body and most still retain TFI so I have plenty of test mules. Lol

[5.0thunder]

Tuffin67, I have created a working TFI trigger. I have only tested it on the bench with a distributor, but once it warms up some I will be using it on the 5.8 in my F250. I added code for the 4 and 6 cylinder TFI as well, but those need more testing.

Chris

How is the testing going with the 8 cylinder setup?

[Jama]

Tuffin67, I have created a working TFI trigger. I have only tested it on the bench with a distributor, but once it warms up some I will be using it on the 5.8 in my F250. I added code for the 4 and 6 cylinder TFI as well, but those need more testing.

Chris

Hey Chris.

I'm looking to finally sit down and have a play with this again as my engine build is progressing to the point the ECU is needed.
Is your latest code somewhere, if so I'm happy to have a play here with it and validate it.

James

[PSIG]

Hey, @Jama - I"ve had to pass on a few conversions that would have been optimal to use a signature tooth decoder, one Holley and three TFI jobs. Any luck with the narrow tooth + wide gap detection? Any info on how Ford, Holley, MS and others do it? Anything I can do to help this get over the hump?

[miker]

Having read this thread Is it as simple as ......

Tracking the rising and falling of the tooth edge (the change setting for the trigger - which means you can only use either hall or optical triggers). From this work out the size of the tooth by the time it takes to complete from low to high and back to low. When you detect a tooth smaller than the rest by a factor of ?? percentage, you've found your trigger tooth. The ?? percentage will have to be calculated every tooth as the engine speed will change with the engine cycle and therefore affect the duration of each tooth.

Do we know if different systems use different sized teeth (or to be more accurate ratio between trigger tooth and normal tooth), different numbers of teeth? are the teeth always equally spaced?
Do we have an idea how fast an engine can change speed (so we can work out if the 'standard' tooth size is within X percent of the last tooth thats ok as its just the engine changing speed, if its greater than X then its the smaller tooth - Ideal percentages to work with are 3,6,12,25,50)

ie can we make it generic so it works with a variety of systems instead of just one?

[PSIG]

Do we know [ ... ]?

ie can we make it generic so it works with a variety of systems instead of just one?

I'd hope so, and TFI/HyperSpark would be the first handy test mule. I know GM, Bosch and others (Citroen?) also used signature tooth, but don't have any spec's. I also assume the wider gap is used to verify the narrow tooth, e.g., <lastTooth AND >lastGap = signature, then tooth-count after that to arrive at #1 again. Maybe. After that it's sync'd and just counting teeth. Again, maybe.

I don't know if all stock signature systems have #teeth = #cyl or extra teeth, but I'd sure like if it can do that, such as using a V8 TFI wheel in a 4 or 6-cylinder application, for better resolution cheap 'n easy. This would assume direct digital signal input (without the module), for control directly in and out of Speeduino. Using the module would require #teeth = #cyl as it will auto-fire on every tooth.

Anyway, if the function is good for the signature, I'm hoping it would be equivalent to missing tooth set-up, and dial-in your number of teeth in TS as-usual.

[EDIT] BTW - if this helps for time, it's from the same page @Jama used for the other images he posted:

TFI_tooth-gap_timing.jpg (48.16 KiB) Viewed 979 times