Timeline for What might be the reason for improper turn on of IGBT
Current License: CC BY-SA 4.0
29 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 2, 2018 at 4:40 | comment | added | vck | @sstobbe yes. It is not properly turning off | |
| Nov 1, 2018 at 13:13 | comment | added | winny | Let us continue this discussion in chat. | |
| Nov 1, 2018 at 13:12 | comment | added | vck | @winny this is someone else's handover to me. So I need to carryforward it | |
| Nov 1, 2018 at 12:34 | comment | added | winny | How much power are you looking at transfering? Have you considered interleaving several boost converters with smaller and better IGBTs instead, to facitiltate the layout? Price could even be lower... | |
| Nov 1, 2018 at 11:39 | comment | added | winny | Oh! minibloc. :-( If you add up the delay, rise and fall times of the device, it's about 450 ns, so your ~900 ns on-time will be just about only switching losses. If you move your scope to your Vge pins, you will see the ground jump and all gate impedance on top of that, so I'm not sure you have anything left. Can you float your scope or use diff probe and place local ground at the collector and measure actual gate on one channel and gate driver output on the other? | |
| Nov 1, 2018 at 11:07 | comment | added | vck | @winny I have added picture which can give little information about routing. Please refer posted circuit diagram along with it. Note- Thickness of PCB is 2.5mm | |
| Nov 1, 2018 at 11:05 | history | edited | vck | CC BY-SA 4.0 |
added 61 characters in body
|
| Nov 1, 2018 at 10:19 | comment | added | winny | @vck Most probably. I understand your struggle if you can't share company proprietary information, but could you sketch how it looks, which layers you are using and what distances and trace widths we are talking about? | |
| Nov 1, 2018 at 9:13 | comment | added | vck | @winny I have added gate-emitter waveform. The shape of this waveform is not looking good. I have observed that PWM at an output of driver IC is clean but at gate-emitter terminal it's shape is totally changed. Is it due to the layout? | |
| Nov 1, 2018 at 9:09 | history | edited | vck | CC BY-SA 4.0 |
added 108 characters in body
|
| Oct 31, 2018 at 12:58 | comment | added | sstobbe | @Vck Your IGBT isn't turning off either. You should be seeing the Collector fly-up after turn-off. What does the VDC supply look like? | |
| Oct 31, 2018 at 11:46 | comment | added | vck | @winny, Vge and Vce waveforms are added in question. | |
| Oct 31, 2018 at 11:41 | comment | added | winny | Understandable. But we still need Vge and Vce oscillogram. | |
| Oct 31, 2018 at 11:33 | comment | added | vck | @winny layout is not possible. | |
| Oct 31, 2018 at 11:30 | history | edited | vck | CC BY-SA 4.0 |
added 68 characters in body
|
| Oct 31, 2018 at 10:38 | comment | added | vck | @analogsystemsrf For security reasons I cannot upload whole ckt diagram | |
| Oct 31, 2018 at 10:37 | comment | added | winny | Please show so. In general, develiping high powered electronics with a two channel scope is doomed to cause problems. Also, show layout. | |
| Oct 31, 2018 at 10:37 | history | edited | vck | CC BY-SA 4.0 |
added 68 characters in body
|
| Oct 31, 2018 at 10:33 | comment | added | vck | @winny I have two channel oscilloscope so i can capture two w/f at a time and as per your advise, I have observed the VCC of the gate driver and it is stable 15V. | |
| Oct 31, 2018 at 10:00 | answer | added | analogsystemsrf | timeline score: 2 | |
| Oct 31, 2018 at 9:54 | comment | added | analogsystemsrf | Show photo of your wiring, including the gate-driver. | |
| Oct 31, 2018 at 8:50 | comment | added | winny | Please show detailed Vge and Vce oscillograms, along with Vcc for your gate driver. | |
| S Oct 31, 2018 at 7:54 | history | suggested | Niteesh Shanbog | CC BY-SA 4.0 |
Fixed grammar
|
| Oct 31, 2018 at 7:02 | comment | added | vck | Yes, thickness of trace is proper. I need opinion about Gate pulse too. Is it appropriate? | |
| Oct 31, 2018 at 6:27 | comment | added | user105652 | @Vick. I mean the IGBT source lead must use heavy gauge wire or wide-thick trace back to power source | |
| Oct 31, 2018 at 5:34 | comment | added | vck | Yes, I am using heatsinkand it is also getting very hot. I didn't understood yoursecond sentense (Source must have a low resistance ground return to power source) | |
| Oct 31, 2018 at 5:16 | comment | added | user105652 | I have found IGBT's to be very sensitive to ground bounce. Source must have a low resistance ground return to power source. Usually 10 to 12 volts will saturate the IGBT ON. Less than 10 volts but > 3V is in the linear region and it will get HOT! Do you have it mounted to a heatsink? You should. | |
| Oct 31, 2018 at 4:59 | review | Suggested edits | |||
| S Oct 31, 2018 at 7:54 | |||||
| Oct 31, 2018 at 4:58 | history | asked | vck | CC BY-SA 4.0 |