Good point, the injector might not be 300V rated.
About the VPN chip, I only see it as a TO220 case. I was hoping we might find one that also includes an SMT package. Those are much easier for me to assemble, but I can do it either way.
Injector Control Options
Re: Injector Control Options
The equivalent DPAK is the VND5N07. It's actually easier to find than the TO220 package.
Jean
Jean
Re: Injector Control Options
Awesome, and thanks. That's exactly what I was looking for. I'd like to keep the SMT or thru hole theme if possible. I guess I should have posted that I was looking for SMT a while back.
I plan to do a bit of a respin of the board. When I do, I'll try to add this SMT version as well. Mor options, same space.
I plan to do a bit of a respin of the board. When I do, I'll try to add this SMT version as well. Mor options, same space.
Re: Injector Control Options
Is it just me or is about the only difference between ST's OMNIFET and OMNIFET II is that the OMNIFET II has slightly slower rise and drop times. I wonder why they called it II.
I see the "Diagnostic feedback through input pin" feature. I suspect there isn't really much we can do with that feature right? I don't think we have enough AN's to capture it, and I'm certain we don't have a software plan to back it. However, I can't help but wonder if that is something we should consider.
I see the "Diagnostic feedback through input pin" feature. I suspect there isn't really much we can do with that feature right? I don't think we have enough AN's to capture it, and I'm certain we don't have a software plan to back it. However, I can't help but wonder if that is something we should consider.
Re: Injector Control Options
AN's? Analog iNput?
Wow, missed one of my favorite threads, and now my head is swimming from reading 4 pages super fast.
Yes, support for 1 ms injection is an absolute necessity. My car idles around 1.2m and I'm not sure if that's including the opening time (I feel it is, meaning it's 200 us of "injection").
I tune this (an autotuner is something I would HIGHLY recommend putting into the code at some point) by using double squirts per cycle, watch the AFR's change, and adjust the opening time. After that, all other tuning is easy. Its iterative, but it works well.
I'm not sure I'm seeing the arguments for hot side verses low side switching, PWM, etc. I thought the plan was hardware defined P&H, not software. If that's the case, we don't need super fast switching. And the injectors open slowly - ~1 ms. So, if the rise time on the voltage DRIVING the coil is 100 us, you've still got tons of room to my thinking. Would it be cool to be able to tune the P&H in software? Yes. But not needed. Being able to read this out and get feedback would be cool, as mentioned much much earlier. I guess I could envision a single software controlled potentiometer (whatever they are called) to tune all the P&H chips... I don't know if that is feasible, I'd guess not.
-Abe.
Wow, missed one of my favorite threads, and now my head is swimming from reading 4 pages super fast.
Yes, support for 1 ms injection is an absolute necessity. My car idles around 1.2m and I'm not sure if that's including the opening time (I feel it is, meaning it's 200 us of "injection").
I tune this (an autotuner is something I would HIGHLY recommend putting into the code at some point) by using double squirts per cycle, watch the AFR's change, and adjust the opening time. After that, all other tuning is easy. Its iterative, but it works well.
I'm not sure I'm seeing the arguments for hot side verses low side switching, PWM, etc. I thought the plan was hardware defined P&H, not software. If that's the case, we don't need super fast switching. And the injectors open slowly - ~1 ms. So, if the rise time on the voltage DRIVING the coil is 100 us, you've still got tons of room to my thinking. Would it be cool to be able to tune the P&H in software? Yes. But not needed. Being able to read this out and get feedback would be cool, as mentioned much much earlier. I guess I could envision a single software controlled potentiometer (whatever they are called) to tune all the P&H chips... I don't know if that is feasible, I'd guess not.
-Abe.
Re: Injector Control Options
Just surfing around the net looking for some info on my JECS subaru ecu. Then on a unrelated serch turned up this little fella.
seems interesting.
http://www.freescale.com/files/analog/d ... 3810FS.pdf
Now sure if anyone else has seen it before.
so i though i would put it up..
here is an eval bord for it.
Interesting to see what they are using for the coil drivers.. i wonder what part numbers they actually are..
http://www.freescale.com/webapp/sps/sit ... 5979968460
I am thinking this is what motec uses, or at least something very VERY similar, in a motec diagnostic page, it can tell you if you injectors are ok, short to ground, open, or under and over current..
I think we(errr, you all) think very carefully about peek and hold drivers. Most OEM turbo engine are running them. And lets face it, any high boosted aplication that is also driven on the road can benefit from them. Yes you can use Jean's board but that is another peice to what could become a rats nest.. How nice would it be to actually have injector diagnostics. Not to mention all the other things that can be accomplished with that chip..
seems interesting.
http://www.freescale.com/files/analog/d ... 3810FS.pdf
Now sure if anyone else has seen it before.
so i though i would put it up..
here is an eval bord for it.
Interesting to see what they are using for the coil drivers.. i wonder what part numbers they actually are..
http://www.freescale.com/webapp/sps/sit ... 5979968460
I am thinking this is what motec uses, or at least something very VERY similar, in a motec diagnostic page, it can tell you if you injectors are ok, short to ground, open, or under and over current..
I think we(errr, you all) think very carefully about peek and hold drivers. Most OEM turbo engine are running them. And lets face it, any high boosted aplication that is also driven on the road can benefit from them. Yes you can use Jean's board but that is another peice to what could become a rats nest.. How nice would it be to actually have injector diagnostics. Not to mention all the other things that can be accomplished with that chip..
Re: Injector Control Options
the biggest board here is the 33810:
one thing that has turned me off about it is that the current sense/analog plane shares the digital plane. I've tried to isolate their plane from my device as best as possible, but I still have testing to perform. Is anyone interested in more info ? anything discrete (capability wise) could not touch this feature set if you value board space (which I do).
as you state, diags are exactly why I chose this - it's also very compact and gives me the choice of which ign driver I want to use. another cool feature is that it can actually measure the spark duration so you can do things like dynamically tweak dwell or possibly detect misfires and then correlate that to crank accel as a doublecheck.
lots of cool things over the last 5 years or so are showing up for the general consumer like us
these are good times!
one thing that has turned me off about it is that the current sense/analog plane shares the digital plane. I've tried to isolate their plane from my device as best as possible, but I still have testing to perform. Is anyone interested in more info ? anything discrete (capability wise) could not touch this feature set if you value board space (which I do).
as you state, diags are exactly why I chose this - it's also very compact and gives me the choice of which ign driver I want to use. another cool feature is that it can actually measure the spark duration so you can do things like dynamically tweak dwell or possibly detect misfires and then correlate that to crank accel as a doublecheck.
lots of cool things over the last 5 years or so are showing up for the general consumer like us
these are good times!Re: Injector Control Options
That's pretty amazing. Any idea what they cost?
Certainly the GPIO capability simplifies things. I'd say put two on the board, and 6 or 8 cyl could use both, everyone else could run 8 extra accessories.
Any idea what they cost?Certainly the GPIO capability simplifies things. I'd say put two on the board, and 6 or 8 cyl could use both, everyone else could run 8 extra accessories.
Re: Injector Control Options
freescale lists them @2.50$usd per/1000.. and there's no possible way you could implement all that functionality in discretes for that cost.
Re: Injector Control Options
i am going to take pics of the motec and the GeMs board for the lancer evo 5 tomorrow at work
but i agree you could never get that functionality from discrete components for the cost. Not to mention the supporting code for it.
I really think it should be looked at..
it is a definite way of separating you from the rest of the MS world and others also..
but i agree you could never get that functionality from discrete components for the cost. Not to mention the supporting code for it.
I really think it should be looked at..
it is a definite way of separating you from the rest of the MS world and others also..