inj. deadtime

[elturbonitroso]

hi guys i found this info of the deadtime of this injectors on here but i dont know what its means and i would like to know the dead times of this injectors at 13.2v.thanks

[elturbonitroso]

nobody

[billr]

I can't speak for others, of course, but that graph posted is gibberish to me, of no value for figuring dead-time or anything else that I can see.

[kaeman]

I don't think I would use that info to set up my engine, it would be better to do the actual testing and input the values into a spreadsheet and get the slope and use that for the deadtime value. much safer course of action.

[elturbonitroso]

ok thanks guys.

[elturbonitroso]

Ok so how accurate it's the old method of changing the numbers of squirts

[subwoofer]

Not very. Doing a proper test of flow rate and dead time only takes half an hour with everything installed in the car. Gives the added benefit of easy upgrades if or when you get larger injectors.

[Yves]

Not very. Doing a proper test of flow rate and dead time only takes half an hour with everything installed in the car. Gives the added benefit of easy upgrades if or when you get larger injectors.

1/2 hour ??

Even with one injector and fiddling with the fuel lines, I can hardly seeing that be done in 1/2 hour.

I've found enough variance in dead times between injectors in one set to warrant testing every and each one of them.

[EspeNS]

Depends on how many parts you have to dissasemble, and how many injectors.
Helped change to bigger injectors on a mates car last summer. They where full of dried E85.
About 1.5hr. That included cleaning the injectors, modifying the clips that hold them against the fulerail, changing one O-ring, the deadtime measurment, and mounting all backt to the engine with the bigger injectors.
4 cyl/4injector engine.

[9secz]

I hope you guys don't see this as an advertisement. I just want to say that I have a very nice injector flow bench that uses a National Instruments Compact RIO controller with an FPGA that provides 25 nanosecond timing resolution. It's very precise! I also wrote the code so that only one injector fires at a time. The controller built in to the flow bench fired all 4 simultaneously, which did not work out too well with 1600 cc injectors.

Shoot me a PM if you want me to test your injectors for match and dead time. I run them at 2, 4, 6, 8, and 10 ms pulse widths and multiple voltages, so it takes roughly 2 hours for a set of 6 or 8. That means I'll ask for money.

Here's a vid of the setup https://www.youtube.com/watch?v=ynCmjip85F0

Jeff

[racingmini_mtl]

I hope you guys don't see this as an advertisement. I just want to say that I have a very nice injector flow bench that uses a National Instruments Compact RIO controller with an FPGA that provides 25 nanosecond timing resolution. It's very precise! I also wrote the code so that only one injector fires at a time. The controller built in to the flow bench fired all 4 simultaneously, which did not work out too well with 1600 cc injectors.

Shoot me a PM if you want me to test your injectors for match and dead time. I run them at 2, 4, 6, 8, and 10 ms pulse widths and multiple voltages, so it takes roughly 2 hours for a set of 6 or 8. That means I'll ask for money.

Here's a vid of the setup https://www.youtube.com/watch?v=ynCmjip85F0

Jeff

Do you only use the VersaFueler for your tests or can you use other drivers? If you can't use the same drivers people are using, in my opinion the results will have little value since there's a very good chance that will not be the correct dead time value. If you can use the same drivers, that would be a useful service worth spending some money.

Jean

[9secz]

Do you only use the VersaFueler for your tests or can you use other drivers? If you can't use the same drivers people are using, in my opinion the results will have little value since there's a very good chance that will not be the correct dead time value. If you can use the same drivers, that would be a useful service worth spending some money.

When driving high impedance injectors, the VersaFueler is transparent because it does not limit initial current increase. It simply grounds the injector the same way the ECU would.

When driving low impedance injectors, it provides 4 amp peak, 1 amp hold current, which is a common scheme for driving low impedance injectors. Here are voltage and current scope traces for a low Z injector driven by a VersaFueler (upper is current, 1 amp per division, lower is voltage, 5 V per division):



If somebody is using something like a resistor box to drive low Z injectors, then yes, I agree with you. Their dead times would be much larger than my measurements, and the dead times would vary quite a bit among the injectors in the set.

Here's a vid of low Z injectors on my bench driven by a resistor box that a guy sent to me for evaluation:

https://www.youtube.com/watch?v=J5E3FqdTSJc

If they are running high Z injectors, or a proper 4/1 peak and hold driver, then my measurements would apply to their setup.

Jeff

[racingmini_mtl]

Jeff,

You're only considering the opening phase. There is another part in the closing time and that means the clamping voltage of the VersaFueler will have an effect on the deadtime. If it is not the same as the one used on the MS injector driver that will skew the measurement potentially significantly. Also, the clamping voltage is not the same for the different injector drivers used in all the MS variants.

Even just considering the opening phase, the voltage drop in the driver will have an impact. So if that is different, you will get different results.

Having said that, it shouldn't be too difficult to modify your test setup to either be able to use different drivers. It would be even better to be able to use the actual ECU and drivers but that would complicate things a bit since you wouldn't be able to use your software solution but would have to rely on the MS test mode.

Jean

[elturbonitroso]

well i did the old method and when i change from 2 squirt to 4 squirt the a/f stay the same and the car start and drive much better but i already order a graduated cylinder to do the test.

[jb007]

well i did the old method and when i change from 2 squirt to 4 squirt the a/f stay the same and the car start and drive much better but i already order a graduated cylinder to do the test.

I profiled my injector including the non linear region and created a spread sheet for the process. I weighed my fuel and knowing the specific gravity it's easy to Calc the volume. Most petroleum companies Web sites list the SG of their fuels. My results came out very close to the flow values of the manufacturer.
See my blog http://www.jamesobartlett.com/?p=201
Jim

[9secz]

You're only considering the opening phase. There is another part in the closing time and that means the clamping voltage of the VersaFueler will have an effect on the deadtime.

Yes, I agree that zener (or whatever you use) voltage will affect the rate of coil current decrease, and therefore injector valve closing time. I use a 33 V zener as you can see from the scope pic I posted earlier in this thread.

I'm not interested in adapting other drivers to my bench, but I am interested in measuring the effect of clamp voltage on dead time. Time to order an assortment of zeners from Digikey.

Jeff

[elturbonitroso]

ok guys how usefull its this info the injectors manufacture send me this.thanks

[elturbonitroso]

[elturbonitroso]

I guess its useless

[billr]

It would be for me; measuring with your drivers and such is the easiest and most accurate way.