For any add-on boards such as VR conditioners, optos and OEM interface boards
User avatar
By cx500tc
#16218
Over in chat the idea was brought up to have some sort of injector controller board to allow use of low-impedance injectors with Speeduino.
A few approaches to that idea were mentioned including using TI's LM1949 Injector Drive Controller, and using a microcontroller-based system control injectors using pulse-width modulation.

So, I'm starting this topic in order to open discussion up to those who, for whatever reason, do not participate in chat.
User avatar
By cx500tc
#16219
Example of a basic LM1949 circuit:
LZ_1949.png
LZ_1949.png (8.3 KiB) Viewed 6043 times
This would be a 1 injector per board configuration. Wiring would be as:
PAD1 > from injector (injector supplied +12V at one terminal, second terminal routes here)
PAD2 > ground (for injector current)
PAD3 > driver enable (apply VCC to enable)
PAD4 > injector trigger signal (active high | VCC)
PAD5 > VCC (+3V to +5.5V)
PAD6 > ground (to ECU)

Potential issues:
- high cost: IC is ~$6 USD each.
- some components may require different values based on injector characteristics to optimize performance.
- uses Darlington transistor to switch injector so consumes a considerable bit of power and can generate considerable heat.
By KLAS
#16225
for me that would be interesting. over here are still a lot of TBI systems using a single Bosch 1.0 ohm injector and i don't like big resistors in series to bring them up to around 14 ohms
By dazq
#16229
I have just created a git at my autohome2 github for this, i will be looking into something simple this week
Along the lines of
Output X ms @ Z duty% then
X ms @ Z duty% until trigger released.
I will base config over a ts project to allow for future changes and adjustment.
Something a bit more complex such as monitoring the current drawn with a output curve graph for duty cycle output can be added after
User avatar
By PSIG
#16230
KLAS wrote:... i don't like big resistors in series to bring them up to around 14 ohms
Why is that? I am not being pointed, but looking for the real problems, so that good solutions can be found.

We need a list of issues and/or goals to these low-Z counterpoints:
Low-Z injectors with drive resistors have been used in millions of cars, are simple, inexpensive, reliable, small in size, obviously work well in those applications, and were specifically chosen for use in the various designs over optional or alternative High-Z injectors (using no resistors).

In addition to the hi-Z list, we need to be sure we understand why low-Z injector/resistor pairs were selected for many designs, in-spite of the (presumably) higher costs when compared to hi-Z injectors without resistors.
:geek:
David
User avatar
By cx500tc
#16240
In general, low-Z injectors open and close more quickly than high-Z injectors, when operated appropriately for their design, i.e. "peak and hold" for low-Z and "constant current" for high-Z. As alluded to, an inline resistor with a low-Z injector will operate much like a similarly flow-rated high-Z injector, given the resistor is chosen appropriately.

Anyhow, I think the open/close rate is one reason low-Z were used on many EFI systems, even with the resistor inline. By selecting the lowest possible value resistor which does not push the drivers (MOSFET or similar) hard enough to overheat, you can achieve an open/close rate quicker than many high-Z injectors of similar flow, which results in more precise fuel metering, even if only marginally more precise.
By dazq
#16242
The lowz injectors i see being used always seem to have large flow rates, eg tbi, cng and lpg ? Is that a pattern?
I noticed that the gas ones in particular tended to be really low sometimes under 1ohm!!
My tbi ones from european hatches run from 4-6ohms.
Unfortunately i dont have a test mule anymore that runs lowz(v8 range rover and fiesta cfi are both no more)
I have no issues trying out a firmware design but aside from bench testing i need somebody to test in the real world....?
User avatar
By cx500tc
#16249
I make no claim as to why some vehicles were done one way and others differently. Some of it may be related to the state of tech available at the time, though. Like my motorcycle calls for 330cc/min low-Z injectors stock, no resistors. ;) Perhaps high-Z just wasn't able to handle the system reliably well, and like many things, some habits are hard to break?

As for continuing on for a time, with low-Z + resistors, perhaps Honda and others had huge stock on the shelves and decided to carry on with what they had, or it just worked and they saw no need to change things until emissions regulations required it?
By Rocket
#16250
Long time ago I have made simple and cheap peak an hold circuit with 555 timer. At mono stable configuration 555 timer give full current for injector 3-4ms to open and for rest time injector stay open by current from resistor. I dont used it a lot because I changed injectors to bigger witch was hi z.
User avatar
By cx500tc
#16251
@psig / David mentioned in chat:
"... a simple tiny PCB with an 8-pin ATTiny uC and a power MOSFET that was snuggled inline of the harness to each low-Z injector resistor. The ATTiny simply activated the MOSFET to bypass the resistor for 0.5ms when the injector channel turned on. Bingo - full amps for 0.5ms to open it, then resisted for the remaining injector time on (injector latency was 0.9ms, with 0.5 open and 0.4 close). Proof of operation came when a JBPerf P&H board was added to the system and the ATTiny boards removed."

So here we have proof a microcontroller could be coaxed into duty for this.
  • 1
  • 2
  • 3
  • 4
  • 5
  • 7

I must have misunderstood what "after few[…]

I've been away from playing cars for a few weeks, […]

I uploaded logs and last tune and dyno plots look[…]

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

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