| Electronics > Projects, Designs, and Technical Stuff |
| LTC3895 step down converter burns off |
| << < (3/6) > >> |
| Miyuki:
--- Quote from: Srijal97 on January 07, 2019, 10:02:50 am ---bootstrap cap has to handle high voltage ripples according to the simulations, that is why we decided to go with the large 250V rated ceramic capacitor. --- End quote --- Then is something horribly wrong Bootstrap cap provide low impedance charge to gate, but have almost no load So where you get high voltage ripples ? It can be tricky as ringing at switching node can easy overcharge it to high voltage and then it blow driver and transistor |
| capt bullshot:
OK, here we go again: The bootstrap cap is totally wrong. One would use a smallish ceramic cap here. There's typically seen 10V ... 15V across this cap, not the full input voltage. At the input node, I'd suggest a SMT ceramic cap, placed near to the MOSFETs (now I can see the real low-side FET, still looks quite small, often the low-side FET has to carry more losses than the high-side one. I didn't check your components. Your GND trace around the positive input to supply the IC is way too long and forms an nice loop that can cause all kinds of trouble, as others pointed out. I'd suggest moving the whole circuitry to the GND side of the power components (from right hand side to left hand side regarding your layout), and use a proper GND plane covering everything from input to output. Avoid vias within GND nets. It's would have been easy to rotate the HS FET within your layout by 90° to get shorter traces to the decoupling caps. For the DCR sensing traces from the inductor: keep them close together. Fill the space between + and - input (in between the pins of C3 / C6) with + input and GND planes, so you can solder SMT ceramic caps across the gap, but keep the electrolytic in place, consider adding another one or two of them. Some fine tuning (within the final circuit) can be achieved by inserting resitors into the MOSFET gate drive lines, by connecting a small value resistor (1R ... 22R range typically) in series with the gates, one can tune the switching times of the MOSFETS to reduce ringing and losses. Anyway, doing so requires a working layout first. Edit: forget about what I said regarding the small low-side MOSFET, I was reading your layout wrong due to the unusal arrangement of + input and GND in respect to the control IC. Anyway, rotating Q3 (which I now see is the low side FET) would be a good idea to shorten its power traces. So, IMO the root cause of the IC releasing the smoke is the poor GND layout, causing transient overvoltages. You might get it working by attaching a as short and wide as possible bodge from Q3 Source to the IC's center pad. But it'll still won't be a good design, better redo the layout. |
| Srijal97:
Yes, the bootstrap was a terrible mistake on our end. Probing both ends of this capacitor shows that the voltage across it is actually just 10V and not the ripple we saw on just one side of it (attached simulation picture showing waveforms across cap). It appears that we will have to get new PCBs done now, but meanwhile, do you think we can perform one more run on the current board with the following changes made? - We'll remove the large bootstrap cap and add a 0805 ceramic close to the IC...I see vias where it can be soldered on the current board itself. - The GND trace problem should be resolved by adding the extra wire as indicated in a post above. Does the existing long trace need to be cut? @capt bullshot, how would small SMT caps be able to handle the 140V input if they're added across the + and - input gap? |
| capt bullshot:
The SMT caps to handle the input voltage won't be small anymore, but still SMT. One can buy ceramic SMT caps rated for such high voltages, though they will be rather large and expensive. You've got a THT cap in your layout, a good film cap will also work here. Don't operate the board without proper input caps, or put some protection circuitry within the IC's power supply. I wouldn't replace the bootstrap cap, it's just ridiculously large but should work. Most important change would be the GND modification. If you have access to such equipment, use a variable output power supply to ramp up the input voltage slowly for the first tests, and observe critical waveforms (e.g. the switching node) with an oscilloscope. Be sure to use a very short grounding path for your scope probe, the usual ground clip supplied with the probe often is too long. |
| capt bullshot:
Here's two examples of power stages / step down regulators I've done to give you some impression of how to do GND planes on a double sided circuit board: http://wunderkis.de/pvbat/converter.pdf http://wunderkis.de/dcdc/Test-LTC3775.brd.pdf http://wunderkis.de/dcdc/Test-LTC3775-2.sch.pdf |
| Navigation |
| Message Index |
| Next page |
| Previous page |