Is the UD2.7C under voltage lock out basically worthless?

[Benjamin Lockhart]

I'd really like to be able to run my DRSSTC from one power cord, but I just tested what happens if the whole coil loses power at once due to the breaker tripping. The under voltage lock out works for a bit, but then stops working again below about 8V! (I tested with a scope on the gates) So for a fraction of a second the bridge is driven with 8V dropping to 0V. This would probably mean death for the bridge if the IGBTs desaturate! Or is it fine?

Do you have any ideas about how to fix this problem?

Also if I just run a separate 30 foot cable for the driver power, will there be any noise problems?

Thanks, Benjamin

[flyingperson23]

I've always used a separate cable for driver, haven't had any problems. Especially if it's a 120v line to a power supply for the driver on the coil, you'll probably be fine. For uvlo, I'd try maybe discharging the bus faster or adding capacitance to keep the UD on longer. On my big coil I used a NC contactor (with 120v coil so it closes immediately) to drain the bus in a few seconds after losing power. You could also just send the gate wires through NO relays so the gates can't be driven unless the UD has full voltage. If you do that, you probably shouldn't leave the gates floating when the relays are off.

I have also a few times in recent memory disconnected my small coil's driver power with full bus power while the coil was on, so if my UD does the same thing as yours that would've driven it at 8v. Hasn't blown up yet.

[futurist]

Hydron is responsible for UD2.7C UVLO design

It's probably easy to find an alternative method to the LM8365 which is both through hole and works better as well.

This particular part was picked due to convenience more than anything else, I wanted an UVLO in my driver and modded the UD2.5 layout to include it, and the LM8365 was a) easy to use with just 3 extra parts and b) I had one in my box of random ICs.
Loneoceans then included and improved it in his UD2.7 design (along with good documentation and other improvements - thanks!).

There is a limitation to doing it the quick and easy LM8365 way - when the 24V supply rail gets really low (e.g. 8V or lower) the chip can stop working, which means the UVLO will stop working and the driver can still try and switch the IGBTs, putting them at risk. To solve this (and enable through hole construction) I would suggest you use a comparator like the LM311 (already in use for OCD), and have it sensing the 24V supply on one input, and a zener or similar reference on the other, placed in parallel with the OCD LM311 output. The LM311 will run down to 3.5V or so, which is low enough that the IGBTs won't be turning on anyway, and can be powered by the 24V supply to make the circuit as simple as one resistor, one potentiometer, a zener diode and the LM311 (maybe plus decoupling cap). See my MSPAINT drawing for an idea of what to do:

[Benjamin Lockhart]

Thanks! In order to implement this on an existing UD2.7C, I would need to remove the existing LM8365 UVLO circuit, correct? Would 50K or so work for the trimpot?

I just found this schematic by Franzoli electronics that has a simple UVLO circuit. Would this also be a good option?

[Hydron]

You are correct that the UVLO has a major flaw at <8V - unfortunately it'd been copied by others before I found the issue. See also another post by me about it: https://highvoltageforum.net/index.php?topic=2054.msg15613#msg15613

[davekni]

For patching existing UD2.7 boards, a couple other options come to mind:

Simplest is to change from LM8365BALMFX27 to LM8365BALMFX45 (with corresponding readjustment of the POT or change from 4.7k resistor to something close to 7.8k).  It is the same part except with a 4.5V threshold instead of 2.7V.  When adjusted for same UVL threshold, low voltage "failure" will be reduced to 60%, so 4.8V instead of 8V.  That might be low enough to avoid IGBTs turning on.

Next option is to scab a 1.25V (or so) shunt regulator chip in series with 4.7k resistor (and either reducing 4.7k to ~2.5k or readjusting POT).  Examples are LM385-1.2, LM4041, LM4051, LMV324, LMV431.

Finally, for a patch that does not require unsoldering anything, it is possible to simplify the diode and resistor solution a little (at a bit more cost):  Add a 3.0V or 3.3V shunt regulator from +5V to LM3365 input (pin 2).  This avoids the need for a resistor divider.  Just replaces the 0.6Vf of a diode with 3V or 3.3V of a reference.  Choices include ZRC330Q, ZXRE330, LM4040D33, LM4040C30.

[Benjamin Lockhart]

I added a 3.3V shunt regulator from 5V to the LM8365 input node and now it only turns back on below about 4.3V which is exactly as expected given that the shunt regulator clamps the input to 3.3 volts below the 5V rail, in other words the input is at 1V when the 5V rail drops to 4.3V.

This is probably fine as the threshold voltage of my IGBTs is 4.5V minimum but that's cutting it pretty close.

Is my understanding correct that a shunt regulator in series with the 4.7k resistor would allow the circuit to function all the way down to the voltage rating of the regulator? In that case the regulator could be 1.25V or 1.5V or 2.5V right, basically anywhere in between 1V and 2.7V?

I just need to make sure the minimum current of the regulator is met at low voltage. 1.25V and 33.5k ohm gives only 35uA. would the LM285LP-1-2 regulator work, it needs only 10uA instead of 65uA like some of the others.

Thanks, Benjamin

[davekni]

I added a 3.3V shunt regulator from 5V to the LM8365 input node and now it only turns back on below about 4.3V which is exactly as expected given that the shunt regulator clamps the input to 3.3 volts below the 5V rail, in other words the input is at 1V when the 5V rail drops to 4.3V.

When 5V is at 4.3V, 24V may be around 7V due to drop through regulator chips.  Still might not be enough to damage anything.

Is my understanding correct that a shunt regulator in series with the 4.7k resistor would allow the circuit to function all the way down to the voltage rating of the regulator?

It will function even lower, down to 1V rating of LM8365 chip, plus a tiny bit more for POT voltage drop as you mention.

In that case the regulator could be 1.25V or 1.5V or 2.5V right, basically anywhere in between 1V and 2.7V?

Best to stay at 1.25V.  By 2.5V shunt regulator, there is only 0.2V from there to LM8365 threshold of 2.7V.  Tolerance on LVO would be scaled from that relatively inaccurate 0.2V difference.

I just need to make sure the minimum current of the regulator is met at low voltage. 1.25V and 33.5k ohm gives only 35uA. would the LM285LP-1-2 regulator work, it needs only 10uA instead of 65uA like some of the others.

LM285 is great, but LM385 is still fine.  What matters is current of shunt regulator at 1V, which is even lower than specified regulation current.  The LM385 that I just measured is slightly under 3uA at 1V.  Adding 0.65 uA for LM8365, 33k * (3.65uA) = 0.12V.  So UVL will work down to 1.12V, 1V from LM8365 threshold plus 0.12V drop across 33k.

[Benjamin Lockhart]

Thanks for your help, that makes sense. I'll try the 1.25V regulator then.

[Benjamin Lockhart]

I finally got around to fixing this UVLO properly.

I removed R30 the 4.7K resistor and added an LM385 1.25V shunt regulator and a 2.7K in series, from the negative rail to TP2 (the LM8365 input).
It works perfectly now with the UVLO activating when the supply drops below 20.5V and not coming back on again at a low voltage at all!
I think now I can safely run my coil with just one power cord.

You might be able to see in the top left where I soldered the 2 parts between TP2 and the negative pad where the bridge rectifier is missing.


Thanks again Dave for your sugestions.

[davekni]

I finally got around to fixing this UVLO properly.

I removed R30 the 4.7K resistor and added an LM385 1.25V shunt regulator and a 2.7K in series, from the negative rail to TP2 (the LM8365 input).
It works perfectly now with the UVLO activating when the supply drops below 20.5V and not coming back on again at a low voltage at all!

I think now I can safely run my coil with just one power cord.

Thanks again Dave for your sugestions.

Great to hear it is working as expected!  You are certainly welcome.