maybe a bank of mosfets ? i will mock up on Monday afternoon my injectors are around 13 ohms if that helps
will update as parts arrive
Excellent. mosfet should be fine. Looks like peak current will be 16V / 13 ohms = 1.23 Amps. For safety plan on 2.5 x 1.25 Amps = 3.125 Amps. (16V assuming automotive injectors)
3 Amp mosfets will do the job and provide a very nice safety margin.
excellent can you work out a simple schematic and i will either mill it out of copper on the cmc or bread board it out and order some n channel mosfets ?
For some ideas on an injector circuit go to the MsExtra sight Megasquirt Support Forum (MSEXTRA) - Index page and click on the manuals link at the top of the page.
It is open source so all of the manuals have the schematics which might give you a pre made circuit.
Just a quick note. I'm up to my eyeballs in work so it will be a week or so before I can provide any kind of schematic. And it will be a rough schematic that I just drop some chips on the board and string a few wires. No review or safety checks until I have more time.
okey well as said i will get to work on the simple frame 
I found this looking up this old project I started a long time ago. I didn't have time or the coding skillset to get it working well then. This was the 3rd thing i ever coded, yes it was repetitive. It did say it wasn't working yet in the first line of comments in the code..
This will likely fill in a lot of the blanks, injector_tester/Readme.txt at master · jrbe/injector_tester · GitHub
The circuitry ideally would drive the injectors with lm1949 chips. Injector Driver Breadboard
Those drive "peak and hold" or saturated injectors properly. Peak and hold means high power initially to open the injector quickly, then maintains about 25% of the peak current to hold it open.
The eventual idea was to have the code fire the injectors sequentially. Giving them their own on and off times was to use the standalone (engine computer) to adjust each injector to balance them to the rest of the set. Injectors can be off 25% to the set, new at low pulsewidths. The repetitive code centoproject found was a first attempt in a project that was put on hold for longer than i wanted.
There were 3 big things that really needed to be fixed besides the excessive repeaty-ness of the code:
-
Fire the injectors sequentially
-
The timing function seemed to need currentMicros - previousMicros to correct the timing drift. (Like https://www.arduino.cc/en/Tutorial/BlinkWithoutDelay but with micros.)
-
Change digitalWrite to digitalWriteFast to help accuracy / speed.
This tester was to be a useful tool for finding injector dead time at different voltages, flow test injectors, find pressure / flow / hammering issues in fuel rails, be adjustable to set a certain pulsewidth (on time) and rpm that was seeming to be problematic while running the engine, etc. Thats why it used micros instead of millis.
mstanley, you did help with a piece of the puzzle.
kvasilak forked my initial code and wrote sequential code but on and off times were the same (50% duty cycle.)
Still a lot of work to get it all together. I still dont have the time to commit to this to make it happen unfortunately.
Just wanted to post this to hopefully keep someone from diving in to a pool that's really not dug yet.
Would like to bring this topic back from the dead as I continue to search for a DIY solution to an adjustable flow test bench.
I have no time to help right now, too many other things going on.
You could look at the speeduino stuff and maybe use their hardware (arduino or teensy 3.5) to run it.
I think megasquirt may have a test mode to be able to use it as a flowbench. With the combo of those 2 this may be done.
I am interested in how you make out, if you get somewhere with it post an update here.