The Lo-D Driver

[rhd]

I think these should be happy in parallel.

LeQ: I think your approach avoids one of the major concerns in running regulators in parallel, which is the risk that individual regs will handle more or less current than each other due to tolerances. I think your approach avoids that possibility. No?

 

[Hiemal]

So I did the test as promised, and

That's 7.9 amps into my little test load (it rose to about 8.1 after a couple seconds) Also, it seems the regulator actually DOES work fine with such little forward voltage, oddly enough, and they shared the current pretty well. This was just with TWO, though, I don't know how more would behave together.

And as far as the problem of one reg handling more current than the other; They should actually self regulate themselves, as they have overtemp built in which automatically lowers the output current if one gets too hot. After a while of battling between eachother I'd imagine they'd reach an equilibrium point and output the same currents together, if that's what ends up happening.

 

[rhd]

Since I would be using a diode on the output of each regulator to reduce the voltage fed to the LD down to 2.0 VDC, I don't need a test confirming they can be ran in parallel, the dropping diode from each regulator would then also be a blocking diode to isolate them from another. I've used diodes in series before to drop the supply voltage by small amounts with other circuits I've built, easier to do that than use a voltage regulator.

I don't think you really understand how this works.

First off, the objective is not to REDUCE voltage output to the LD, it's to RAISE voltage of the overall load so that the current sensing circuitry works. Absent the diode, it's not as if your regulator gives the LD more voltage. These are constant current drivers, so your LD is going to drop the same voltage regardless of the presence of a diode in series with it.

Second, the diode does not isolate the drivers in a manner that would prevent differences in tolerance from causing relatively more or less current to be supplied by each. In fact, if you intentionally mis-set one to 4A, and one to 3A, they would remain out of balance in this setup, regardless of your diodes, because these drivers are using independent high side current sensing. Unless I'm mistaken, but I don't think I am.

 

[Alaskan]

15W C-mount 400um FAC IR diode for burning.

RHD, I would have the values set to be as close as I can get them so all of the cc regulators would produce close to the same current, of course there would be some which pull a bit more, due to small differences. The way I view constant current devices, they produce a voltage, the voltage is either higher or lower depending upon where the current is set to be, current can only flow with a voltage present into a load or resistance. Set the current higher and it can only produce more current if the voltage is raised higher, set the current to be lower and the voltage must be lower or nothing different happens. CC's just measure the amount of current flowing as referenced by a voltage drop anyway, right?

The only reason I would have another diode in series with the LD is so that the cc regulator can operate at a higher voltage output on the circuit board itself, if it can't operate at 2 VDC output or that low of a voltage output. Using a dropping diode in series would allow the circuit board to operate where it can be turned up to work at a higher range (if it won't in the lower range) to produce over 2 volts output and the dropping diode then reduces the voltage to the laser diode due to the voltage drop across it, the current would be set to the same regardless and the actual voltage out of the cc regulator would take care of itself, always following to produce the set current. That's how I'm seeing it, just a simple work around for cc regulators which cannot work with low forward voltage biased laser diodes because they stop working when their output voltage is that low.

 

[Hiemal]

Well regardless, it appears the regulators work fine with only 2 volts of forward voltage ANYWAY, so it shouldn't be needed. Basically the reason why I was suggesting the extra series diode in the first place, is because there's a differential op amp on the board, a ZXCT1009 to be specific. This has a minimum voltage requirement of 2.5 (according to the datasheet). IR diodes have a forward voltage of 2 volts, roughly, which is below what should be the minimum, but it seems in practice it still works just fine. The extra series diode would have been added to the laser diode's forward voltage, making it roughly 2.5 - 2.7 volts.

But again, it doesn't appear to even be needed.

Your explanation of cc sources is correct, yes.

 

[rhd]

Alaskan - the problem is, you keep saying "the dropping diode reduce the voltage to the laser diode". That's not correct. Adding the dropping diode does not reduce the voltage to the laser diode. The voltage across the laser diode will be exactly the same, regardless of whether you add a dropping diode or not, because the voltage across the laser diode is a product of the current, and the current is identical in both scenarios.

 

[rhd]

Well regardless, it appears the regulators work fine with only 2 volts of forward voltage ANYWAY, so it shouldn't be needed. Basically the reason why I was suggesting the extra series diode in the first place, is because there's a differential op amp on the board, a ZXCT1009 to be specific. This has a minimum voltage requirement of 2.5 (according to the datasheet). IR diodes have a forward voltage of 2 volts, roughly, which is below what should be the minimum, but it seems in practice it still works just fine. The extra series diode would have been added to the laser diode's forward voltage, making it roughly 2.5 - 2.7 volts.

You may be forgetting to factor in the voltage drop of your RSENSE. Unless you've intentionally optimized your board for minimal sense loses (by sticking to real close to the 50mV sensing threshold), you may be dropping several hundred mV across the RSENSE. Add that to some drop across your leads, and you may be at 2.5V.

 

[Hiemal]

You may be forgetting to factor in the voltage drop of your RSENSE. Unless you've intentionally optimized your board for minimal sense loses (by sticking to real close to the 50mV sensing threshold), you may be dropping several hundred mV across the RSENSE. Add that to some drop across your leads, and you may be at 2.5V.

The sense resistor is 0.015 ohms. At max current that should give a Vdrop of about 63 mV (since 4.5 amps x 0.015ohms) so that's not it either...

 

[Alaskan]

RHD, I think I might understand your point, putting a diode in series with the output would mean the cc regulator would just raise the voltage output that much more to achieve the current it is set to output, so self-defeating?

 

[Hiemal]

RHD, I think I might understand your point, putting a diode in series with the output would mean the cc regulator would just raise the voltage output that much more to achieve the current it is set to output, so self-defeating?

Nope.

The theoretical reason why adding a diode in series with the LASER diode, is so the forward voltage of what the driver needs to supply goes up. Why make the forward voltage higher?

Because the current sensing op amp theoretically needed more voltage to properly regulate everything. That's all the extra diode is for, nothing more nothing less. However in practice it doesn't seem to matter, and the driver regulates fine with 2 volts of forward voltage.