beast. 50 cents.

If the output tries to swing much below the + rail, it blows up!

The output swings to within 0.6 volts of either rail, which does not suggest darlington followers.

The PNP can saturate. But the composite can turn on to within one junction drop of the - rail.

John

John, please, I know you're smarter than that. Have you actually constructed a section of it? I know there are no values to go by however, experienced engineering helps out when it comes to that.

If you would like, I have an LTspice of that output section in working order.

The whole idea of that circuit is to hold constant current between the two poles on the output. Using the proper diode forward voltages you don't get very much current and you need to place a load on the output to actually see the effects.

The lower output only turns on hard if there is a (+) voltage at the output bridge that is not coming from the upper rail pair. This is to clamp the output, for example in the case if you were to operate a device from the high side and use this device as a sink or, it could simply drain off unwanted junk. The fact there is a PNP at the bottom allows the bottom to turn on a lot harder when needed when this connection takes place.

This whole design is actually very good and keeps a nice shoot through on the circuit at all times.

Like I said, the offer is open.. I only constructed enough of it to show on side of the wave but that should be enough to explain it I'll keep this in my files for a while.

Jamie

I would like. Please.

The compensation node (the one with the caps on it) can never be more than 4 junction drops below V+. How can that work?

John

Schokky Diodes..

Jamie

John S wrote:

Version 4 SHEET 1 880 680 WIRE 320 -192 -272 -192 WIRE -272 -144 -272 -192 WIRE -272 -48 -272 -64 WIRE 320 -48 320 -192 WIRE 320 -48 -32 -48 WIRE 432 -48 320 -48 WIRE 320 16 320 -48 WIRE 320 16 240 16 WIRE -32 64 -32 32 WIRE 176 64 -32 64 WIRE 320 64 320 16 WIRE 432 64 432 -48 WIRE -32 96 -32 64 WIRE 256 112 240 112 WIRE -32 192 -32 160 WIRE 320 192 320 160 WIRE 432 192 432 128 WIRE 432 192 320 192 WIRE 320 272 320 192 WIRE 688 272 320 272 WIRE 320 336 320 272 WIRE 320 336 224 336 WIRE 688 352 688 272 WIRE -32 384 -32 320 WIRE 96 384 -32 384 WIRE 160 384 96 384 WIRE 320 384 320 336 WIRE -224 432 -304 432 WIRE -96 432 -144 432 WIRE 256 432 224 432 WIRE 96 448 96 384 WIRE 224 448 224 432 WIRE 688 480 688 432 WIRE -32 496 -32 480 WIRE -32 512 -32 496 WIRE 320 512 320 480 WIRE 224 544 224 528 WIRE 96 560 96 512 FLAG 320 512 0 FLAG 224 544 0 FLAG -32 576 0 FLAG -272 -48 0 FLAG -304 512 0 FLAG 688 480 0 FLAG 96 560 0 SYMBOL npn 256 64 R0 SYMATTR InstName Q1 SYMATTR Value 2N2222 SYMBOL pnp 160 432 M180 SYMATTR InstName Q3 SYMATTR Value 2N2907 SYMBOL npn 256 384 R0 SYMATTR InstName Q4 SYMATTR Value 2N2222 SYMBOL res 208 432 R0 SYMATTR InstName R1 SYMATTR Value 100k SYMBOL schottky -48 256 R0 SYMATTR InstName D1 SYMATTR Value 1N5817 SYMATTR Description Diode SYMATTR Type diode SYMBOL schottky -48 192 R0 SYMATTR InstName D2 SYMATTR Value 1N5817 SYMATTR Description Diode SYMATTR Type diode SYMBOL res -48 -64 R0 SYMATTR InstName R2 SYMATTR Value 100k SYMBOL npn -96 384 R0 SYMATTR InstName Q5 SYMATTR Value 2N3904 SYMBOL res -48 480 R0 SYMATTR InstName R3 SYMATTR Value 4.7k SYMBOL voltage -272 -160 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value 10 SYMBOL Misc\\signal -304 416 R0 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V2 SYMBOL res -128 416 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R4 SYMATTR Value 56k SYMBOL schottky 448 128 R180 WINDOW 0 24 72 Left 2 WINDOW 3 24 0 Left 2 SYMATTR InstName D4 SYMATTR Value 1N5817 SYMATTR Description Diode SYMATTR Type diode SYMBOL pnp 176 112 M180 SYMATTR InstName Q2 SYMATTR Value 2N2907 SYMBOL schottky -48 96 R0 SYMATTR InstName D3 SYMATTR Value 1N5817 SYMATTR Description Diode SYMATTR Type diode SYMBOL res 672 336 R0 SYMATTR InstName R5 SYMATTR Value 10 SYMBOL cap 80 448 R0 SYMATTR InstName C1 SYMATTR Value 10p TEXT -328 -216 Left 0 !.tran 0 .01 0 startup

Excuse my sloppy editing. This is just the output section that is in question.

Jamie

It's only swinging low because you're yanking it to ground with a 10 ohm resistor. Try returning the load resistor to +5.

John

There is a complete model for the TCA0372 if you google hard enough. I think I have it somewhere.

I just took some educated guesses for components.

Jamie

I don't see anything wrong with your components (for a spice run), but the output is half-wave rectified. Have you seen that?

John S

wn.

ng

the standard way of driving high ohm fuel injectors is an open drain with a voltage clamped fet, diode clamping to supply makes it close far too slow

-Lasse

Why do you care? If it meets the specs, I *like* it. Got some on order to play with.

Good find.

John S

Because I'm a circuit designer.

John

I know, LH=Linear Hybrid.

He was 50 when you were in your second year at MIT, so he might have beaten you to something.

Wow, LH4009 at 10,000V/us.

The other 2 are in my 1980 book.

Ok, It seems to be opening and closing ok as it is now but I am open to suggestions.

We first fully saturation the coil for each spray cycle then operate the coil at a variable current (PWM) to control the volume. We found this necessary because the injector was exhibiting unreliable smooth starts. It may have something to do with the chemical we are injecting. We have experimented with Bosch EV1 and EV6 models now. The EV6 seems to be the best. We also experimented with throttle body types and got a socking surprise with the amount of current required there.

I understand the concerns of back current in the coil holding a magnetize field however, since the current flow is changing direction wouldn't there be a short moment of repulsion and therefor pushing the actuator in the opposite direction (core Hysteresis). By the time this energy dissipates the actuator should be on it's way to the inactive position ? Just a thought how ever, I did notice that we had better control over the valve using that diode once we enter the partial current mode.

Onset of that, any form of clamping the energy is going to cause current to be exerted in the coil during collapse, I mean, how else would you deal with it other than letting it out which isn't very healthy for your components and it also likes to generate R.F.

I suppose I could also use a snubber across the fet to slow down the collapse and keep the voltage with in tolerable range.

My thoughts on the matter of energy release is why not capture it for reuse which is another problem because this unit may end up on UPS power in the event of power failure. So I am dabbling with the idea of placing a half bridge on the high side of the coil where I can switch off the + lead and clamp it to common, at the same time I will be employing the wheeling diode back to the +supply just before the H-bridge which will have a battery on the system as part of the UPS.

---- Back to that chip ---

Since that chip in question does not seem to show ratings for that diode, at least I didn't see it? I can only assume it's not intended to be used as such..

Jamie

they are "normal" high impedance injector, like used in almost all cars. I'e never seen them driven in any way other than with a self clamping fet like for example, VND7N04 . full on to open it, let it flyback to the clamp voltage to close it

injected volume it always controlled by how long it is on

They might be low impendance injectors, they need a high current to open and then a lower to stay open, they are used because they generally open and close faster so the injected volume is easier to control oover a wider range

you'll need a peak-n-hold driver, more way to do that

current is still flowing in the same direction

with out the diode the current drops fast so the pwm cycle would have to be very fast, but that is not how you should be driving them they should be fully on or off

the clamped fets are designed for it, datasheet specs clamping voltage and the allowable energy per pulse

guess you could do that, though they energy involved is hardly worth it

opening and closing time would be close to the same, since charging and discharging voltage for they coil would be similar (same if you clamp at 2*supply)

with no spec probably not

-Lasse

There are some 10K-class opamps around, but most have pitiful voltage swings. THS3201 is pretty good, 10,500 v/us and +-7.5 volt supplies.

John

Then change Fig1 so it looks the way the specs says it works.