Decent Chinese UD Works well except for phase lead not working.

[DJthefirst]

    When building my coil I noticed that my UD2.7s output was not working correctly so I decided to try one of the Chinese UD boards. Images are attached below. The drive works really well and I was able to translate all the documentation except for the driver works fine when I use it with Phase lead disabled but if it is enabled I get no activity from my IGBTs. The Driver has a Phase lead enable switch (S1) and a Phase lead compensation switch (S2) which just increases the indcutance. ITs on the left-hand side of the schematic are attached below (IDK how to insert images into a msg). The GDT signal looks the same for both cases other than it has no feedback from OC bc the IGBTs aren't firing. Any ideas why this could be.

Video of coil working.

TLDR: Driver works with Phase lead disabled but in both cases, the GDT puts out an 18v signal.

[davekni]

With D3 and D4 being normal signal diodes (1N4148) on this board, it will take a stronger CT feedback signal to start oscillation compared to UD2.7 (which uses MBR0530 schottky rectifier diodes with lower forward voltage drop for its corresponding D1 and D2).  You might get enough initial feedback signal by biasing bridge outputs with bleed resistors, one side to VBus- and the other side to VBus+.  Try both ways around (swap + and - sides).  One way will give the biggest initial half-cycle of current.  Scope CT output (driver input) to see the initial half-cycle.

There are other patches that could get around this problem if above doesn't work.

[thedoc298]

    When building my coil I noticed that my UD2.7s output was not working correctly so I decided to try one of the Chinese UD boards. Images are attached below. The drive works really well and I was able to translate all the documentation except for the driver works fine when I use it with Phase lead disabled but if it is enabled I get no activity from my IGBTs. The Driver has a Phase lead enable switch (S1) and a Phase lead compensation switch (S2) which just increases the indcutance. ITs on the left-hand side of the schematic are attached below (IDK how to insert images into a msg). The GDT signal looks the same for both cases other than it has no feedback from OC bc the IGBTs aren't firing. Any ideas why this could be.

Video of coil working.

TLDR: Driver works with Phase lead disabled but in both cases, the GDT puts out an 18v signal.

Hello DJthefirst, Im getting ready to start testing the same board. Spent several days back and forth trying to see what the jumpers and switch's did. I understand most settings but  never understood what they were trying to say about  the little 2 pin jumper that is Under where is says UD2.9 DRSSTC Drive, most I got was some kind of protection. Will be testing first of the week and will let you know how it goes. Thought about just switching the d3 and d4. Also is it necessary for anything on the "temp control enable"  or the "start enable"..? That is what they called anyway. 


PS: I just seen the .txt file, that is nice. Also thanks davekni for you tip, going to test out first. Thanks

[davekni]

Im getting ready to start testing the same board. Spent several days back and forth trying to see what the jumpers and switch's did. I understand most settings but  never understood what they were trying to say about  the little 2 pin jumper that is Under where is says UD2.9 DRSSTC Drive, most I got was some kind of protection. Will be testing first of the week and will let you know how it goes. Thought about just switching the d3 and d4. Also is it necessary for anything on the "temp control enable"  or the "start enable"..? That is what they called anyway.

I'm guessing the "little 2 pin jumper" is "S3" in the schematic.  It appears to be an internal pulse width and duty cycle limit, auto-cycling in case the optical enable input lasts longer than ~200us.  I don't see anything for UD2.9 pulse-skipping mode in the schematic.

This schematic looks to be a fairly-close match to UD2.7, with a few extra features, and the slight problem of 1N4148 for D3 and D4.  Using AC08 instead of HC08 for the signal path is nice.  Less delay requires less phase lead to compensate.  (Fast logic does require good ECB layout and supply bypassing.)  The S3 duty-cycle limiting could also be helpful in case of interrupter control errors.  C31 and/or R15/R16 values may need tweaking to match your coil's needs.  Or skip this by inserting S3 jumper.

How much did this board cost?  May be easier and/or cheaper than finding all the BOM parts for UD2.7 these days.

[thedoc298]

Im getting ready to start testing the same board. Spent several days back and forth trying to see what the jumpers and switch's did. I understand most settings but  never understood what they were trying to say about  the little 2 pin jumper that is Under where is says UD2.9 DRSSTC Drive, most I got was some kind of protection. Will be testing first of the week and will let you know how it goes. Thought about just switching the d3 and d4. Also is it necessary for anything on the "temp control enable"  or the "start enable"..? That is what they called anyway.

I'm guessing the "little 2 pin jumper" is "S3" in the schematic.  It appears to be an internal pulse width and duty cycle limit, auto-cycling in case the optical enable input lasts longer than ~200us.  I don't see anything for UD2.9 pulse-skipping mode in the schematic.

This schematic looks to be a fairly-close match to UD2.7, with a few extra features, and the slight problem of 1N4148 for D3 and D4.  Using AC08 instead of HC08 for the signal path is nice.  Less delay requires less phase lead to compensate.  (Fast logic does require good ECB layout and supply bypassing.)  The S3 duty-cycle limiting could also be helpful in case of interrupter control errors.  C31 and/or R15/R16 values may need tweaking to match your coil's needs.  Or skip this by inserting S3 jumper.

How much did this board cost?  May be easier and/or cheaper than finding all the BOM parts for UD2.7 these days.

I paid around 85 bucks and have it mostly mapped out, lot of nice features and mostly like what was said, mostly ud2.7. Will buy a second soon as I get this firing.

[DJthefirst]

    I paid 90 for mine to get it from eBay instead of AliExpress. For temp enable and start enable just short them. I think it is designed for connecting to external soft-start controller and temp sensors which are really nice features I'm planning to use down the line. Overall I think my UD2.7 was ~110$ for all the parts so I think this board is a real help to the community. In the txt I just went through with google translate and used some of my college friends to translate the schematic which has nothing important on it that wasn't said in the post. At the bottom was the summary.
 
    I will also try and scope out my CT to see the voltage difference between Phase Shift on and off. For biasing the FB do you think it would hurt to connect the Vbus to the Driver through a resistor? I've been trying to keep logic isolated from Power bc of arcs and my Variac doesn't actually turn off it keeps my V Bus +/- referenced to 120vac. It was surprising when I hooked my scope up and tripped the breaker after connecting the ground and everything was off. I'll also try to replace the diodes I think I have extras laying around from my ud2.7

[davekni]

I will also try and scope out my CT to see the voltage difference between Phase Shift on and off. For biasing the FB do you think it would hurt to connect the Vbus to the Driver through a resistor? I've been trying to keep logic isolated from Power bc of arcs and my Variac doesn't actually turn off it keeps my V Bus +/- referenced to 120vac. It was surprising when I hooked my scope up and tripped the breaker after connecting the ground and everything was off. I'll also try to replace the diodes I think I have extras laying around from my ud2.7

With phase shift off, you should see roughly almost +-5V square wave on FBCT input.  Instead of 51-ohm burden resistor + inductor, CT load is 4.3V zener + schottky diode pairs.

I wouldn't recommend any resistors from VBus to driver.  Usually VBus is referenced to line power, and driver referenced to safety ground.  If you are thinking about my previous suggestion for adding resistors, I was recommending resistors from bridge outputs to VBus, nothing to driver.  A resistor between the two H-bridge outputs is common.  Two separate resistors, one from each output to VBus, provides twice the initial current half-cycle amplitude.  Works only if outputs are tied to the "correct" VBus polarity (one to VBus+ and the other to VBus-).  Need to try both ways around, as one gives double starting current and the other way gives no starting current.

With a variac at 0V, both outputs are tied to one side of line voltage.  Usually that is neutral, which should be roughly at ground potential.  I'd check your variac and house line wiring to see if neutral and hot(line) are swapped.  (If the breaker is GFCI, then even a neutral-to-ground short will likely trip it.)  Still not a good idea to connect scope ground to anything line-referenced unless line connection is unplugged or fed through an isolation transformer.

BTW, another way to minimize required starting FBCT current is to add a 215k resistor (220k is close enough) from comparitor - input to +5V.  May be easier to patch to D3 anode or D4 cathode or right side of R5, since those parts are all wired to comparitor - input.  Adding that resistor will bias the comparitor inputs close together when quiescent, even closer than changing diode parts.

[thedoc298]

Im getting ready to start testing the same board. Spent several days back and forth trying to see what the jumpers and switch's did. I understand most settings but  never understood what they were trying to say about  the little 2 pin jumper that is Under where is says UD2.9 DRSSTC Drive, most I got was some kind of protection. Will be testing first of the week and will let you know how it goes. Thought about just switching the d3 and d4. Also is it necessary for anything on the "temp control enable"  or the "start enable"..? That is what they called anyway.

I'm guessing the "little 2 pin jumper" is "S3" in the schematic.  It appears to be an internal pulse width and duty cycle limit, auto-cycling in case the optical enable input lasts longer than ~200us.  I don't see anything for UD2.9 pulse-skipping mode in the schematic.

This schematic looks to be a fairly-close match to UD2.7, with a few extra features, and the slight problem of 1N4148 for D3 and D4.  Using AC08 instead of HC08 for the signal path is nice.  Less delay requires less phase lead to compensate.  (Fast logic does require good ECB layout and supply bypassing.)  The S3 duty-cycle limiting could also be helpful in case of interrupter control errors.  C31 and/or R15/R16 values may need tweaking to match your coil's needs.  Or skip this by inserting S3 jumper.

How much did this board cost?  May be easier and/or cheaper than finding all the BOM parts for UD2.7 these days.

I paid around 85 bucks and have it mostly mapped out, lot of nice features and mostly like what was said, mostly ud2.7. Will buy a second soon as I get this firing.

I have five pictures with instructions, but could only send one, is that the limit

[davekni]

I have five pictures with instructions, but could only send one, is that the limit

Yes, you can add more, but it may not be obvious how.  Click on the "(more attachments)" line below the existing attachment file selection box.

Nice to see confirmation of my previous 2-pin jumper explanation, ~200us enable pulse width limit (protection).

[thedoc298]

I have five pictures with instructions, but could only send one, is that the limit

Yes, you can add more, but it may not be obvious how.  Click on the "(more attachments)" line below the existing attachment file selection box.

Nice to see confirmation of my previous 2-pin jumper explanation, ~200us enable pulse width limit (protection).

Nice call, nice to have someone to verify what Im not real sure about.

[davekni]

The first quote from Capture2.JPG:  "1.As shown in the diagram, this switch is used to select whether to turn on anti-dept compensation. The current picture shows that the compensation function is turned off.  It is necessary to use the compensation function to start to the other side (please switch all switches in the case of power failure)"

Do you know which switch this is in the schematic?  Is this phase-lead (S1 and/or S2), or perhaps some other switch that is missing from the schematic?  Not critical for me to understand, as I don't have this board.  Might be useful in answering future questions, though.

[DJthefirst]

    I believe he is referring to the phase lead enable but the translation is off. There are no other switches on my board at least. I fixed my ac problem the connectors on the variac were mislabeled.
    I also added the resistor across the comparator and now I can tune phase lead but it made things kind of worse. My signal is 90* out of phase and definitely hard switching. I can shift the signal but only by a small amount and the amplitude switch didn't seem to make too much of a difference. 1st is with phase shift disabled, Second is phase shift centered, Third all the way left, Fourth all the way right. I have some inductors from my ud2.7 but I don't want to make too much of a mess of my board and the phase didn't change much anyway. any solutions or recommendations?..

[thedoc298]

The first quote from Capture2.JPG:  "1.As shown in the diagram, this switch is used to select whether to turn on anti-dept compensation. The current picture shows that the compensation function is turned off.  It is necessary to use the compensation function to start to the other side (please switch all switches in the case of power failure)"

Do you know which switch this is in the schematic?  Is this phase-lead (S1 and/or S2), or perhaps some other switch that is missing from the schematic?  Not critical for me to understand, as I don't have this board.  Might be useful in answering future questions, though.

This is my hacked note's, if you can read it.

[thedoc298]

The first quote from Capture2.JPG:  "1.As shown in the diagram, this switch is used to select whether to turn on anti-dept compensation. The current picture shows that the compensation function is turned off.  It is necessary to use the compensation function to start to the other side (please switch all switches in the case of power failure)"

Do you know which switch this is in the schematic?  Is this phase-lead (S1 and/or S2), or perhaps some other switch that is missing from the schematic?  Not critical for me to understand, as I don't have this board.  Might be useful in answering future questions, though.

[thedoc298]

The first quote from Capture2.JPG:  "1.As shown in the diagram, this switch is used to select whether to turn on anti-dept compensation. The current picture shows that the compensation function is turned off.  It is necessary to use the compensation function to start to the other side (please switch all switches in the case of power failure)"

Do you know which switch this is in the schematic?  Is this phase-lead (S1 and/or S2), or perhaps some other switch that is missing from the schematic?  Not critical for me to understand, as I don't have this board.  Might be useful in answering future questions, though.

I posted a picture below, sorry for the clutter, of the two switches the one to the right is the on or off and the sw to the left is the one that goes to half the inductor.

[davekni]

I believe he is referring to the phase lead enable but the translation is off. There are no other switches on my board at least. I fixed my ac problem the connectors on the variac were mislabeled.

I posted a picture below, sorry for the clutter, of the two switches the one to the right is the on or off and the sw to the left is the one that goes to half the inductor.

Thank you for the clarification.  Translation is often problematic.

I also added the resistor across the comparator and now I can tune phase lead but it made things kind of worse. My signal is 90* out of phase and definitely hard switching. I can shift the signal but only by a small amount and the amplitude switch didn't seem to make too much of a difference. 1st is with phase shift disabled, Second is phase shift centered, Third all the way left, Fourth all the way right. I have some inductors from my ud2.7 but I don't want to make too much of a mess of my board and the phase didn't change much anyway. any solutions or recommendations?..

Which two signals are displayed in your scope traces?  I'm guessing the lavender trace is Vge on one IGBT and blue is FBCT at the driver input terminal.  If that is correct, I suspect the driver schematic isn't accurate, or there is a ground-loop issue with scope probing.  FBCT input can't have 12V peak with 4.3V zeners to ground (phase lead off).