One of my Zen Lights has died! - Page 8 - diyAudio
Go Back   Home > Forums > Amplifiers > Pass Labs

Pass Labs This forum is dedicated to Pass Labs discussion.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 15th September 2012, 06:52 PM   #71
diyAudio Member
 
Zero Cool's Avatar
 
Join Date: Sep 2004
Location: MN
Ok we are gettin somewhere! I added a 150 ohm gate resistor right at the mosfet and with the bench supply made up of a transformer a bridge and a 10kuf cap no more blown mosfets! BUT connect up the ZL supply and blammo dead mosfet!

SO, i am trying the network as suggested made up of a 9.1K and 47K resistor with a 47uf cap to ground in place of r2. should i put the 47K on top connected to the + supply? or the 9.1K on top??
__________________
It's not about what you can BUY...It's about what you can BUILD!

Last edited by Zero Cool; 15th September 2012 at 07:00 PM.
  Reply With Quote
Old 15th September 2012, 07:00 PM   #72
diyAudio Member
 
Zero Cool's Avatar
 
Join Date: Sep 2004
Location: MN
with the 47k on top we have another dead mosfet!
__________________
It's not about what you can BUY...It's about what you can BUILD!
  Reply With Quote
Old 15th September 2012, 07:13 PM   #73
diyAudio Member
 
Zero Cool's Avatar
 
Join Date: Sep 2004
Location: MN
Dead Soldiers
Attached Images
File Type: jpg 2012-09-15_13-07-59_295.jpg (409.4 KB, 87 views)
__________________
It's not about what you can BUY...It's about what you can BUILD!
  Reply With Quote
Old 15th September 2012, 08:23 PM   #74
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
9.1K on top, 47K to mosfet, cap in between to ground, but the fact that the mosfet blew anyway probably indicates there is more going on here. I'd check the SOA rating of that mosfet both dc and over a 10msec and 100msec interval and make sure you are not exceeding it. Are you sure that initially you are not exceeding the rated Vds of these mosfets?

Had a recent interesting SOA problem with a mosfet based inrush suppression circuit, turns out that the original design grossly exceeded SOA during the first 100msec of operation. (Voltage across the device combined with the initial current through it, oddly enough this was a 120A rated, 150W part, but the SOA allowed no more than a few hundred mA with >25V across it, and at a Vds of 1V 100A or more was OK) So check that something hasn't changed causing you to exceed some device maximum rating. (Higher average line voltage, etc..)
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 15th September 2012, 08:30 PM   #75
diyAudio Member
 
Zero Cool's Avatar
 
Join Date: Sep 2004
Location: MN
Well the supply is 55v no load and about 45v under load. the 044 mosfets are 55v parts but I also have 140 mosfets that are 100V parts and they blow too!

BUT, if I lower the variac so that the supply is at about 35v under load. they seem to survive. so maybe the CLC in the ZL supply is causing a spike at turn on?

Zc
__________________
It's not about what you can BUY...It's about what you can BUILD!
  Reply With Quote
Old 15th September 2012, 08:36 PM   #76
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
It is quite possible, the 55V parts probably would not survive under high line conditions, but the 100V parts should provided that their SOA is not exceeded during power up. One of the things I would investigate would be whether or not the CLC rings when the FET actually starts to turn on.. Should such be the case a snubber across the inductor might or might not help, a small resistance in series with the choke might help too by lowering the effective Q of the inductor.

Might be worth simulating the supply in PSUDII..
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 16th September 2012, 01:40 AM   #77
diyAudio Member
 
Zero Cool's Avatar
 
Join Date: Sep 2004
Location: MN
Thank you for all the help!
__________________
It's not about what you can BUY...It's about what you can BUILD!
  Reply With Quote
Old 16th September 2012, 01:09 PM   #78
diyAudio Member
 
Join Date: Nov 2009
Location: Silicon Valley, California
Are the broken devices shorted from drain to gate?

The maximum voltage stress from drain to gate is no more than 20 volts (per data sheet). Actual devices would probably not breakdown until stress reached > 30 volts, but probably < 55 volts.

With input shorted to ground, initial voltage stress from drain to gate is 55 volts. When input is not shorted to ground, initial voltage stress drain to gate is 40 volts.

You were close when you used zener to limit voltage, but you needed to limit drain to gate voltage, not gate to source voltage.

One solution while shorting input, is to use 27v zener drain to gate.

Another solution is to place resistor in series with gate. Not 100 ohms, but more like 20K ohms.

Did this help?
  Reply With Quote
Old 16th September 2012, 01:35 PM   #79
diyAudio Member
 
Zero Cool's Avatar
 
Join Date: Sep 2004
Location: MN
All the dead devices are shorted all three legs together! everything shorted! just dead..plain dead!

AHHHH Drain to gate Zener the one thing I didn't try. When the new tube of mosfets get here I will try that and see how it goes. I did notice with the 150 ohm gate resistor that on switch flip power up with the bench supply that it was more stable. less bouncing around. the D-S voltage would always do this sort of slow wave before settling in and with 150 ohms at the gate it would just snap right in and be stable right away. But the ZL supply does this very fast bounce which I think is the 2nd bank of caps charging through the slow inductor. which a better soft start system will help once i get this other issue fixed.

20K gate resistor? I have never seen a gate resistor that high in any amp before. 1K in some ampeg bass amps is as high as i have ever seen. normally i see low values, 47ohm to 150ohm. I'm willing to try 20K but will that affect operation? with 2000pf of gate capacitance will 20K affect HF perf??

Thank you for the help also!


Zc
__________________
It's not about what you can BUY...It's about what you can BUILD!
  Reply With Quote
Old 16th September 2012, 03:46 PM   #80
diyAudio Member
 
Join Date: Nov 2009
Location: Silicon Valley, California
Your failure mode is gate oxide rupture (blown). Short from gate to drain will also measure as gate to source and as gate to substrate. Too much voltage at turn-on from drain to gate. It only takes < 100 nanoseconds for gate oxide to breakdown. This occurs at ramp-up of power supply voltage. When you slowed power supply ramp up, input capacitors (C2 & C5) starting charging from 0v towards 15 volts. As gate charged, drain to gate voltage stress stayed under breakdown voltage.

There is no practical way to measure breakdown voltage without actually breaking down the oxide, which does not recover. However, measurement of one device from a tube will be indicative of other devices in tube. In my experience, we always allowed a 50% margin, but usually less than 100%.
The lower the gain of the device, the thicker the oxide and thus, the higher the breakdown voltage.

The theory of the high gate resister - actual gate will be isolated from 100uF input capacitor, which is initially discharged, by the high gate resistance. Internal 1000pF gate to drain capacitance will charge internal 1000pF gate to source capacitance to ~ 1/2 of drain to source voltage. Of course, as gate voltage goes up, device turns on, reducing drain to source voltage, current starts flowing and power supply voltage goes lower.

Is 20K too high? - 20K and input capacitance of 2000pF (plus internal miller capacitance of ~6 x 1000pF) will limit high frequency response.

Have I made you an expert on MOS devices yet?
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Papa's Zen Lights are no slouch! Zero Cool Pass Labs 8 13th May 2012 10:02 PM
My Rotel just died hybridfiat Solid State 4 1st August 2008 02:33 AM
M DCX died, what to do? sqlkev Digital Line Level 3 22nd May 2006 08:03 PM
Need help--powered sub just died mouse Subwoofers 3 29th October 2003 03:18 PM
Has my monitor died? meta Parts 2 11th December 2002 07:03 PM


New To Site? Need Help?

All times are GMT. The time now is 04:47 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2