Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

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 10th September 2012, 04:39 PM   #111
tomchr is offline tomchr  United States
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Greater Seattle Area
Quote:
Originally Posted by Xpersephone2 View Post
I have been trying to build the design from this thread for a target output of 500V at 500mA approx. for a KT88 parallel push pull amplifier and have come up against many problems.
As far as I can see any linear regulator using recent MOSFET designs other than a very few specific devices will probably fail if the pass element exceeds as few tens of watts dissipation at any time during start-up.
As you point out, you have found the SOA limits of your MOSFETs. I was in the same boat when I started building regulators. You can dig through the earlier pages in this thread for more detail.

Basically, the issue is that many loads - in particular capacitive ones - present a short circuit on the output of the regulator at start-up. This causes tremendous power to be dissipated in the cascode MOSFET, which in many cases leads to the premature death of an expensive MOSFET.
The current limiter does help, though. However, the cascode will need to handle the full power of Vin * Ilimit for a short amount of time. This is where you run into problems with high voltage, high current designs. I spent quite a while pouring over SOA curves in data sheets and simulating in LTspice before settling on the STW... device I list in the BOM.

So yeah... What you describe is completely normal in high power designs. Welcome to the club...

In your case, I would suggest running two regulators, 250 mA each. That would allow you to set the current limiter lower, giving the MOSFETs a better chance of survival.

Quote:
Originally Posted by Xpersephone2 View Post
I've also now build the 21st Century Maida design with the IXYS IXTK22N100L and it works fine too starting up into difficult loads.
Thanks for sharing. It's good to hear that you were able to make it work.

~Tom
__________________
21st Century Maida Regulator, Universal Filament Regulator, Damn Good 300B, Novar Spud, and more...
Neurochrome : : Audio - http://www.neurochrome.com/audio - Engineering : : Done : : Right
  Reply With Quote
Old 19th September 2012, 01:28 PM   #112
diyAudio Member
 
Join Date: Jan 2010
Location: Germany
Hi Tom,

Iīm wondering about the wattage of R1/R3 and R9 found in the latest schematic from 26th march 2012.
R1 ( 100K/2W ) will "see" Vin-Vout-10V. To reach round about 2W on 100K UR1 must be 447V. I canīt imagine when this can happens.
For R3 (15R/2W) a current of 365mA is need to reach 2W at R3.
R9 (200K/3W) will need not more than 1W because on 430V Vout you will get only 0.92W.

For my point of view only R3 need to be a "big" resistor with 2W if you realy want to flow current of more than 250-350mA.

Can you explain why you dimension R1 and R9 in such a way?

Thanks a lot
Karsten
__________________
/// Never forget your towel \\\
You who are doomed, enter and abandon yourself to despair!
  Reply With Quote
Old 19th September 2012, 02:19 PM   #113
diyAudio Member
 
bayermar's Avatar
 
Join Date: Oct 2007
Location: Tuebingen
Hi Karsten,

he might just be cautious and rates the resistors conservatively.
I try to overrate resistors by 2-3 times of the actual dissipated power - this way they stay reasonably cool.
With the max. allowed power, resistors tend to get unpleasantly hot - in fact way too hot for most capacitors that might be nearby.
I guess this is what Tom did, too, but you might want to hear it from the horses mouth (so to say).

Best,
Martin
  Reply With Quote
Old 19th September 2012, 04:51 PM   #114
tomchr is offline tomchr  United States
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Greater Seattle Area
Quote:
Originally Posted by karsten21 View Post
Iīm wondering about the wattage of R1/R3 and R9 found in the latest schematic from 26th march 2012.
R1 ( 100K/2W ) will "see" Vin-Vout-10V. To reach round about 2W on 100K UR1 must be 447V. I canīt imagine when this can happens.
For R3 (15R/2W) a current of 365mA is need to reach 2W at R3.
R9 (200K/3W) will need not more than 1W because on 430V Vout you will get only 0.92W.

Can you explain why you dimension R1 and R9 in such a way?
Most modern power resistors are made from materials that can tolerate very high temperatures. That's how a 2 W modern resistor can be the size of a 1 W or 0.5 W from "back in the day". If the resistor is dissipating the rated power, it will get screaming hot. As in 200~250 deg C. That's too hot for comfort in my opinion.

In my setup, using the values indicated in the schematic, the resistors typically heat up to around 80~100 deg C (measured with an IR thermometer). This with the regulator sitting out in the open on a piece of plywood in my living room (don't do this at home kids!)

I tend to go with a 4~5x safety margin on power resistors. I might go as low as 3x if the 4~5x gets me into ridiculous resistor wattages.
Just for reference, aerospace and military designs generally use a 10x derating factor. They've done a lot more math on the subject than I have. Just sayin'...

R1 is a 2 W type, because during start-up into a capacitive load, this resistor actually sees the full input voltage across it (well, technically V(R1) = Vin - Vzener, but Vzener is tiny in comparison to Vin). To handle this peak power, a 2 W type is needed, even though it only sees a few tens of mW during normal operation.

Hope this answered your questions.

~Tom
__________________
21st Century Maida Regulator, Universal Filament Regulator, Damn Good 300B, Novar Spud, and more...
Neurochrome : : Audio - http://www.neurochrome.com/audio - Engineering : : Done : : Right

Last edited by tomchr; 19th September 2012 at 05:01 PM.
  Reply With Quote
Old 20th September 2012, 06:35 AM   #115
diyAudio Member
 
Join Date: Jan 2010
Location: Germany
Hi Bayermar, Hi Tom

Thanks for your explanation. To sum up the choosen wattage of the power resistors is done for robustnes and lower temperatur only.
Yepp, seemed to me the right way but unfortunately I have only 80x100mm PCB size for the whole PSU including a EL48 transformer. So I must be carefully choose size and place of potential hot devices.
Well, my PCB will not be a part of an aerospace but why not to try to do it in the same way

Thanks a lot for your hints,

regards
Karsten
__________________
/// Never forget your towel \\\
You who are doomed, enter and abandon yourself to despair!
  Reply With Quote
Old 20th September 2012, 11:19 AM   #116
diyAudio Member
 
Join Date: Jan 2010
Location: Germany
Hi folks,

First try of routing a 80x100mm PCB. Itīs designed to feed only 15-20mA on 200 - 300 V for small preamps. Its only single sided and use also a few SMDīs. If it works I will provide you the eagle .brd and .sch schematic

Have fun,
Karsten
Attached Images
File Type: jpg 21th-Maida.jpg (149.8 KB, 546 views)
File Type: jpg 21th-Maida-sch.jpg (72.0 KB, 541 views)
__________________
/// Never forget your towel \\\
You who are doomed, enter and abandon yourself to despair!
  Reply With Quote
Old 20th September 2012, 02:35 PM   #117
dgta is offline dgta  United States
diyAudio Member
 
Join Date: May 2011
Tom, what is the dropout of this regulator? Have you tested it at high voltage, high current and low Vin-Vout? Like on the order of 10V? I don't see anything obvious beside the LT chip dropout and series (limiter) resistor, but I thought I'd better check with the designer
  Reply With Quote
Old 20th September 2012, 04:29 PM   #118
tomchr is offline tomchr  United States
diyAudio Member
 
tomchr's Avatar
 
Join Date: Feb 2009
Location: Greater Seattle Area
I have tested with 400 V, 225 mA out and dialed the variac on the mains input down to the point where the regulator lost regulation. This tends to happen around 410~415 V DC in a setup with a few volt ripple. So the drop-out is on the order of 10~15 V. I recommend that people design for a worst case drop-out of 20 V to have some margin. The spreadsheet I have on my website does all the necessary math for you.

The drop-out is mostly determined by the zener bias, D2, on the cascode.

~Tom
__________________
21st Century Maida Regulator, Universal Filament Regulator, Damn Good 300B, Novar Spud, and more...
Neurochrome : : Audio - http://www.neurochrome.com/audio - Engineering : : Done : : Right
  Reply With Quote
Old 19th February 2013, 03:00 PM   #119
diyAudio Member
 
Join Date: Nov 2002
Location: Cologne, Germany
Can anyone who actually built it comment on the sound of this ? Maybe compared to shunt-regs or maxed-out passive systems ?
  Reply With Quote
Old 19th February 2013, 03:57 PM   #120
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
If you move the three film caps and two protection diodes to the right a bit, you have room for a connector for B+
  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
A Linn Isobarik PMS clone for the 21st century? ingenieus Multi-Way 68 19th August 2013 07:52 PM
Maida Regulator PCB SpreadSpectrum Tubes / Valves 98 21st July 2012 07:25 AM
Maida regulator Andrewc556 Tubes / Valves 21 31st December 2010 03:29 AM
21st century tube? Non-Vacuum Trout Tubes / Valves 4 1st February 2009 02:18 PM
Maida regulator billr Tubes / Valves 15 23rd September 2007 03:19 PM


New To Site? Need Help?

All times are GMT. The time now is 09:10 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright Đ1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2