Quantcast
Channel: Power management forum - Recent Threads
Viewing all articles
Browse latest Browse all 35901

UCC27714: HB,HS,HO leakage failure

$
0
0

Part Number:UCC27714

Three phase motor 12.5kHz PWM commutation 100VDC cause random failure to HS/HO junction, no signs of under/over shoot HO via soft turn on/off compensation resistors.

The means of using soft or even hard NFET turn on always results with high side Totem pole failing, stresses HS/HO internal NFET. Such results in leaking VS supply onto 1/2 bridge inductor raising typical 12.6v floating bias to nearly 14.3v, e.g. HB leakage via HS/HO breakdown.

Conditions: 80us overlapping PWM drive signals HI/LI inputs, LO/HO:130 ohm GTon, 75 ohm GToff (parallel R/Schottky), NFET Qg=<96NC  @VBpin=13.6v

Contrast: Other vendor 600v gate driver (IGPK +/-600/350ma <10us pules) never fail similar described leakage, mandates unwarranted chip replacements.

1. How can maintaining IGPK +/-4amp (<10us pulses) ever cause HO/HS stress via 130 ohm GTon 75 ohm GToff resistor?

2. What is work around to stop HS to HO reverse leakage path from developing in first place, typical HS pin R value?

3. Why does HO/HS leakage path failure cause uncontrolled and very high inductive voltage overshooting?  

4. Zener diode (15.2VzMin, 16.4VzMax @10ma) or 20k protection resistor placed across HS/HO, regulates floating VB the cause? 

5. Any Wiki report updates how to correct HO side gate drive failures caused from industry typical configurations?


Viewing all articles
Browse latest Browse all 35901

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>