For anything you'd like to see added to Speeduino
By GMC270
Would it be possible to add peak and hold injector control to speeduino?

It would be very nice for use with low-z injector from e.g. GM TBI or LPG gas (as in vapour) injectors.

Especially for oldtimers I like to use TBI since it looks similar to a carburettor. You can still have the original look of the engine but have full control of the mixure.
In countries where gasoline is expensive, oldtimers often run on LPG.

I know you can add resistors on low-z injectors and use simple on/off pulses but doesn't that decrease injection accuracy?
User avatar
GMC270 wrote:
Sun Jun 17, 2018 2:42 pm
I know you can add resistors on low-z injectors and use simple on/off pulses but doesn't that decrease injection accuracy?
In my experience with GM TBI, Ford CFI, and other low-Z TBI setups; both PWM P&H and resistor setups work equally well. In-fact, on one TBI system I used used both, and the tune did not change, indicating there was no functional performance or accuracy difference between PWM or P&H (MS1Extra KiethG code) over the resistors I used (MS1Extra 5Ω/25W wire-wound) on the same injectors, engine and tune.

My TBI preference is for the mid-'80s Ford CFI, as it looks more like a carburetor for classic conversions, fits standard carb manifold bolt patterns adn adapters (no special TBI adapters), and has a much wider selection of injector sizes and types to choose from as the injectors are standard Bosch EV1-style. This means you can also use saturated (High-Z) if you prefer. ;) The GM appears more often in articles and web sites as it is a bit easier to find, and was the first to have the ECM hacked, and so became the one people quote without knowing any alternatives; and as we all know, the Internet is powered by parrots repeating info without first-hand knowledge or verification. That has all changed now with home-brew EFI systems, making the CFI a better choice for me. I have not tried converting a CFI or TBI to LPG injectors :?: , so I'm curious about that if you go that way.

If you want P&H drivers, other aftermarket driver boards such as the JBPerf will work with Speeduino, and there are a couple different designs being worked-on here. However, the priority and effort is currently low, as there is no apparent benefit other than adding one fairly expensive board to remove two cheap resistors. ;)


What CFI looks like fresh from the wreckers:
Ford CFI.jpg
Ford CFI.jpg (13.16 KiB) Viewed 1864 times
By noisymime
I'll try not to rant on this topic :D

In my opinion, 95% of setups will never see the difference between a dedicated P&H driver and resistors. The only cases where it really makes a critical difference are where you're right up against the limits of your injector/s anyway (95+% DC). In pretty much all other cases, it just slows the opening time down to around what you get with high-z injectors, which can very easily be tuned for without any negative side affects.

Lots of people have different views on this, but I've not seen any hard data that has shown any different to the above
By GMC270
thanks for the info. If you guys compared low- and highZ injectors one to one without problems, then using resistors will not have a big disadvantage. Makes me wonder though why OEMs use the PWM. to save a couple of cents on resistors? to get the absolute max out of the injectors?
Anyway, thanks for the info!
User avatar
GMC270 wrote:
Thu Jun 21, 2018 5:27 pm
Makes me wonder though why OEMs use the PWM.
While there may be reasons to use PWM or other P&H in specific cases; globally, most OEM systems use resistor packs/boxes/blocks. ;) It is most likely that saturated/high-Z injectors are common for the same cost and simplicity reasons, as although they are generally a bit slower, they also avoid both resistors and P&H circuits. We often see features that may seem "better", but are often done for reasons we do not know, such as cost, or emissions, or the preference of the engineering manager or accountant that day. I do know of one case that specific injectors were used, only because those were the ones they could get the best price on, with the specific spray pattern and flow they wanted for their application.

While I understand your concerns, using a GM TBI with stock injectors on a stock engine, I finally ran out of time when trying to fire 8-squirts per-rev, and it began to lean-out at a bit less than redline. The point here is that the speed of the injector (in most cases) is not a concern itself, as long as it does not limit RPM with too much time spent in opening and closing each rev; thus low-Z is common for multi-shot TBI and high-revving engines like motorcycles. Fueling resolution with large injectors at idle/cruise is a separate issue, and again time is not a big factor in itself, other than the latency (open time or dead time) must be consistent and the setting correct for accurate fueling in any case. You're the engineering manager today, so use what makes you giggle.
By Tjeerd
I converted a carb engine to TBI and single point LPG injection using a microsquirt. I used 4 normal multipoint LPG injectors injecting LPG just before the throttle body. And the TBI of an 2.5 liter doge ram van for petrol injection. So 5 low Z injectors in total.

I can tell you that you do not want to use multi point LPG injectors as a single point LPG injection system. You need a way to " smooth" the injection pulses created by the system. If you don't you get a very uneven fuel distribution between cylinders. (similar to the siamese port problem for EFI and 4 cylinder engines, google it and see). TBI solves this more or less by dumping liquid fuel on the manifold that evaporates over a period of time. Thereby smoothing the pulse. This does not work for LPG because it is vapour injection.

The resistor approach may worked for the TBI injector (did not try that). But it did not work very well for the LPG injectors. I think the LPG injectors did not like the resistor because the big difference between peak (4A) and hold (1A) current needed for that injector. So i changed the approach and used the CS453 4.4A Injector Solenoid Driver. These are all in one Peak and Hold drivers in a to220 package. I use one driver for two LPG injectors (2.2A peak per injector, not enough according to spec). But this works very nicely. Also used the same chip for the TBI injector.
By GMC270
Very interesting Tjeerd, how did you solve the problem with "smoothing" the lpg pulses? did injecting sequentially help compared to batch injection?
By Tjeerd

For the moment its batch only (using one injection channel for 4 injectors) table1 is used for this channel. The other channel is used for the TBI for patrol (using table2). Switching between fuels is done by a relay powering the LPG or the Petrol injector.

We designed and 3d printed an adapter from TBI to air filter. This has some channels inside to try and smooth the flow a bit. Now we can drive the car but it is still not perfect. The outer cylinders valve must be adjusted very quick probably because they are running lean still.
User avatar
+1 and why the recent code mods to increase injections from 1 or 2 per-cycle to infinite can be important for some installations. That was the purpose of my testing with TBI on gasoline, in order to test real-world results of the cylinder-to-cylinder fuel distribution. It works very well, and effectively provides each cylinder intake stroke with its own fuel pulse. With the divided cylinders paired in alternating revolutions, only 2-per cycle per injector (2 squirts on 4 cylinders per injector) gave good results with liquid gasoline, but 1-per cylinder would be necessary with gaseous fuels for the reasons Tjeerd stated. These concepts fall to a number of different induction systems from TBI to Siamese to pulsed water injection systems.

As to the injectors, typical gasoline types have no issues on resistors including versions that use over 6A P&H opening current (CFI) or over 11A (TBI). But if the design and operation specs of your injectors require a different profile, you may need special drivers that can provide it. Given Tjeerd was applying 2.2A max for opening LPG injectors, and I was applying 2.6A@14V on TBI injectors with resistors, it may be worth some experimentation to see if they are responsive to a minimum resistor current as an alternative. @Tjeerd - what were the injector spec's and resistor values you tested?

@GMC270 - I don't mean for us to hijack your thread in any way, and we can drop the injector conversation or move it to another thread if you prefer.

By GMC270
@PSIG, no problem, it is very interesting what you are discussing about and still on topic. Please keep the discussion in this thread! :D

@Tjeerd, how did u check if you are running lean on the outer cylinders? With temperature sensors in the exhaust for each cylinder? Do u see a better distribution when running on the TBI?
Another question (since I was having the same idea with the LPG injection :) ), how do you compensate for the varying temperarure and pressures in the LPG vapour fuel rail? I know it can vary a lot... But AFAIK Speeduino has not enough inputs and no software function for fuel rail pres/temp compensation.

I've been looking into the barometric corrections,[…]

E190 2.3 8V FUEL ONLY

Head cover painted special heat color

The term you're looking for is I/O Expander. Here'[…]

That's weird. Even if my fork is somehow corruptin[…]

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