ud 2.7c poped caps

[costas_p]

Hello to all,

I finished soldering the ud2.7c smd version from loneocean.

I substituted FDD8424H with AOD609 part

I did not install IFD95T neither OPF2412T fiber receivers (as they have not arrived from ordering yet), and connected the controller to 24v power supply to check the voltage rails, as loneocean is suggesting in the first power up of the board (without GDT and CTs connected)

Upon powering up, all indicator leds went up, and after a couple of seconds the SMPS i was using went to current limiting mode, capacitors C22, C23 ,C28, C29  (4.7uf X7R 50v) popped and the 4x AOD609 had overheated upon inspection by touching.

As far as i can understand, not installing IFDT95T and having the R11 (1k resistor) sends a constant on signal to the 74hc14 ic, however loneocean says on his website, that without GDT the controller should be quiet (not operating the output stage ?).

I checked with a microscope, no solder bridges were found, also D9, D10, D11 signal diodes had a strange discoloration.

I guess the failure was because i failed to install the optic receivers, however any one has a better clue ?

[Mike]

All the symptoms you list are consistent with the power supply being connected backwards, how sure are you that it was the correct way around?

[costas_p]

I desoldered all components after phasing jumpers, and the rest of the board powering as expected. I also double checked all diode polarities

[davekni]

capacitors C22, C23 ,C28, C29  (4.7uf X7R 50v) popped

Did these capacitors crack?  Did they break into pieces when unsoldered?  Not sure exactly what "popped" means for a ceramic capacitor.  Damage isn't obvious in posted image.

Did you have the phase-lead jumper SV1 installed, in either position?  Only possibility that comes to mind is that TL3116 is oscillating at a high frequency.  Enable is continuously true without opto receivers, so any oscillation is passed through driver chips and output FETs.  Oscillation in the MHz range seems consistent with many of your symptoms.  If SV1 jumper is not installed, then I can imagine such parasitic oscillation of TL3116.  If SV1 jumper is installed in either position, then current through R27 should bias TL3116 inputs far enough apart to prevent such oscillation.

Good luck with debug and repair.

[costas_p]

After desoldering the ceramics, they were cracked in the middle of them, SV1 jumper was installed in IN possition

Good luck with debug and repair.

Thank you

[davekni]

After desoldering the ceramics, they were cracked in the middle of them, SV1 jumper was installed in IN possition

High frequency oscillation is still the only cause I can think of that fits your symptoms.  Seems unlikely with SV1 installed.  Perhaps there is some other component that is bad.

If you have a scope, I suggest checking signals at phase jumpers before reinstalling any parts.  (And check +9V regulator output.)  If phase jumper signals are quiescent, one high and one low, then install driver chips w/o output FETs.  Measure driver outputs.  Again one output should be high (+9V) and the other low (0V).  For this and subsequent tests, it would be best to have a lower current limit on supply.  If adjustable bench supply is available, that is great.  If not, something like small low-voltage incandescent light bulb(s) in series with 24V input can suffice.  I cut sections of old Christmas minature-light strings.  100-light strings are often available for $1 after Christmas.

Of course, once your opto receivers come, running tests at low enable duty cycle helps limit part damage.