DFH - Defacto FreeEMS Hardware in KICAD

[Fred]

Tell me the desired version and I'll put a note in the readme about it :-)

[AbeFM]

OMG this is killing me! So, I have a megasquirt II provided switched 12V source.




If I don't want to blow the 5V input, I need to clamp the upper voltage. If I put an inline resistor, I need to keep the current over 20 mA and under 120. But what's the voltage drop? I guess a volt and a half.

If I make a divider, I don't know how much current is sucked off by the thing.

All in all it's a mess. Someone give me a good R1 and R2 to get this going.

[jharvey]

There are a bunch of things I don't know about the above. I've got a MSII in the basement, however, I'm banned from the documentation that goes with it, so I don't know what's going on with it.

I would guess, that you need more current sink capabilities, perhaps you found a Vce issue, or your silicone was getting hot. You want to drive it harder, to make sure you're saturated and keep Vce down also keeping your heat down. It may be possible you're exceeding the devices limits.

Remember transistors are current multipliers, and the gate to emitter will be around .7v max. This means that Q7 (I guessed it at Q7) will have to conduct about .006 amps before the 120 ohm starts to leak current through Q6. I believe the impedance of Q6ge will be a couple hundred ohms, and with out current limiting Q7, you will drag the V at Q6ge up to 5V at about .040 amps, perhaps more or less depending on the impedance of Q6ge. Once you drag up Q6ge you will stop conducting though Q7. Also you really don't want to pull Q6ge away from .7V, if you do you are simply over driving and will likely break the gate.

I think you really want to keep a similar setup as what you originally had, just stick the two transistors in a Darlington setup. Perhaps change your drive resistors to change your drive current, as a Darlington setup will have a much higher B, and won't need nearly as much drive current to make it turn on.

Some other notes about Darlington transistor setups can be found here.

http://www.kpsec.freeuk.com/trancirc.htm
http://www.technologystudent.com/elec1/transis2.htm

They also make Darlington transistors, that come in a 3 pin device. You could also replace your existing transistor with a Darlington transistor, if my above guesses are correct that is.

[AbeFM]

Wow, after seeing your response, I have to actually apologize for wasting your time. I gave NO up front explanation.

The source signal is 0 or 12 V digital

I want to drive a 12 V signal (actually, it's a switched ground), but at a higher current than the original signal may reliably support.

I already have a darlington, and it's supposed to be between 20 ma and 120 ma for safe, full saturation operation (across the base (gate?)).

I want to know how to keep it in saturation but not blowing up. From what you said, it sounds like I take 12V - (0.7V x 2) and figure out what resistance I'll need to hit my current, put it inline and I'm done. All I have to do is pick a value where I can have 10V or 17 and it should work in all reasonable cases.

Sound right?

[jharvey]

I'd say that sounds reasonable, but I'm also still flying blind on much of the overall circuit. For example, just because a CPU pin can drive 20ma doesn't mean it won't do it with out some resistance. So your 12V signal may be less than 12V. I might also guess that the 12V signal isn't a CPU pin, so there may be other effects caused by the drive circuit.

A 120mA drive for a Darlington sounds a bit high. Typically a normal transistor has B=100, and Darlington's multiply that, so often it's around B=1000 or B=10000 for Darlington setups. So 120mA would get you at least 120 amps, but more like 1200 amps.

You probably don't need the second drain resistor, but it's also not a bad idea. That drain resistor is real important for MOSFET's and IGBT's because they are voltage devices, and they can hold themselves on when under load because the silicone die works as a resistor divider. If you push enough current through the silicone, it can create a potential on the gate, that keeps it on. The drain resistor bleeds of that little bit of voltage induced by the dies resistance, and allows it to shut off. You can get similar effects with transistors, but it's a lot less common. I don't think you'll have a problem in your application, but it also doesn't hurt to have it there.

How are you getting the 12V? Your vehicle voltage may range from as low as 10.4V (with a dead battery at startup) up to 14.4V (I've seen trucks go up to 16V). You would likely want your current to work out for that range. Also remember that a 7812, requires at least 13V to make 12V. So 12V into a 7812 will make more like 11V.

[SleepyKeys]

I think I understand what you are looking for. Here I go.

My first post to pass information around was these pictures.

The top page is here http://i324.photobucket.com/albums/k352 ... MS_pg1.gif
The power reg is here http://i324.photobucket.com/albums/k352 ... S_pg30.gif
The injector supply is here http://i324.photobucket.com/albums/k352 ... S_pg21.gif
The ADC filter is here http://i324.photobucket.com/albums/k352 ... MS_pg7.gif

Then I attached this

http://www.diyefi.org/forum/download/file.php?id=23

Then these

http://www.diyefi.org/forum/download/file.php?id=31
http://www.diyefi.org/forum/download/file.php?id=30

Then
http://www.diyefi.org/forum/download/file.php?id=33
http://www.diyefi.org/forum/download/file.php?id=32

then
http://www.diyefi.org/forum/download/file.php?id=36
http://www.diyefi.org/forum/download/file.php?id=35

then
http://www.diyefi.org/forum/download/file.php?id=38
http://www.diyefi.org/forum/download/file.php?id=37

then
http://www.diyefi.org/forum/download/file.php?id=40
http://www.diyefi.org/forum/download/file.php?id=39

then
http://www.diyefi.org/forum/download/file.php?id=50
http://www.diyefi.org/forum/download/file.php?id=49

then they come from Sourceforge
http://sourceforge.net/project/showfile ... _id=618983

Do you need the Sourceforge links? I think those might be easier obtained from SF.

Also I see I posted the history at SF as well, so perhaps we should just use the SF for the GIT history.

Are these your most recent diagrams? I wanted to have a look at the injector driving stuff.

Thx,
Sean

[SleepyKeys]

There are a bunch of things I don't know about the above. I've got a MSII in the basement, however, I'm banned from the documentation that goes with it, so I don't know what's going on with it.

Get a refund!! Initiate a charge back etc.
-sean

[AbeFM]

I'd say that sounds reasonable, but I'm also still flying blind on much of the overall circuit. For example, just because a CPU pin can drive 20ma doesn't mean it won't do it with out some resistance. So your 12V signal may be less than 12V. I might also guess that the 12V signal isn't a CPU pin, so there may be other effects caused by the drive circuit.

Ah, it's a stepper motor driver on the MS-II card, should drive like 600 mA, not worried.

A 120mA drive for a Darlington sounds a bit high. Typically a normal transistor has B=100, and Darlington's multiply that, so often it's around B=1000 or B=10000 for Darlington setups. So 120mA would get you at least 120 amps, but more like 1200 amps.

120 is the current not to exceed! 20 is the number I want to be well above.

You probably don't need the second drain resistor, but it's also not a bad idea. That drain resistor is real important for MOSFET's and IGBT's because they are voltage devices, and they can hold themselves on when under load because the silicone die works as a resistor divider.

That's interesting, I didn't know that. Since this is going to be PWM, I may very well want to ensure it shuts off quickly.

Then again, the darlington has internal resistors as I've shown on the right of the drawing ("equiv"), you think that's enough?

How are you getting the 12V? Your vehicle voltage may range from as low as 10.4V (with a dead battery at startup) up to 14.4V (I've seen trucks go up to 16V). You would likely want your current to work out for that range. Also remember that a 7812, requires at least 13V to make 12V. So 12V into a 7812 will make more like 11V.

It's the switched power of the car (yes, 10 to 16) - but run all around inside the MS which is what I'm trying to isolate with this circuit.

Now at least I feel like I'm getting somewhere. Seems like a 10.4 - 1.4 = 9V => 450 ohm resistor would give me 20 ma, and only be 32 mA. If I use the more common 330R I get 44 ma at 16V and 27 which works quite nicely. Do you still think I should drop the drain resistor?

[jharvey]

Are these your most recent diagrams? I wanted to have a look at the injector driving stuff.

Most recent posted are in the SF repo, I've got a PDF version up there as well that can be viewed with out installing special software.

I have made some very small changes that haven't been posted yet. I changed a typo, added a comment and broke a ground path between the drive silicone and regulator ground.

Pages 10-15: Pull down resistor is now 10K.
Pages 25-26, 28: R94 or equiv is just a place holder for folks that might want snubbing capabilities. I added a note to help clarify.

Things I plan to change, but haven't gotten to it yet.

I think RPM input should change to this (changing the LM chip to Max chip)



I also want to add SMT versions of the injector and ignition drive silicone. I'm not sure they will fit.

The AN and Digi Protects have high resistance values, they should probably drop to 1K or less. That can be done imperially, or if some one makes a more refined recommendations, no complaints. The PCB layout is expected to stay the same for this change.

Generic digi drives are low side drive, I should change them to high side drive.

[jharvey]

Get a refund!! Initiate a charge back etc.

DIYautotune offered to honor a refund, even though I didn't have a receipt. DIYautotune is top notch as far as I'm concerned. A refund would just hit him in the $ department, and not change MS. Why punish him for Lances actions. I got the MS unit as a gift (that's why I don't have a receipt), so changing it for $ didn't seem like a right thing for me to do.