OPUS 5.0 A Modern Mullard - Page 4 - diyAudio
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 23rd February 2010, 04:41 AM   #31
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
Quote:
Originally Posted by Eli Duttman View Post
TOTALLY unnecessary! Being majority carrier only devices, Schottky diodes do not exhibit a reverse recovery spike.
But they exhibit higher capacitance than conventional diodes, and it's internal capacitance and parasitic circuit and package inductance what produces RF ringing (LC tank resonator, which needs external RLC with custom values for optimum damping).

btw: A complete waste to use a SiC diode in place of a 1N4007. Some of us are actually doing commercial switching power electronics and we think twice before using SiC due to the high cost... In places where it really improves performance... And the MBR40250... I use them in the power supply of an amplifier capable of 10kw peak, so it's funny to see it in a 20mA supply
__________________
I use to feel like the small child in The Emperor's New Clothes tale

Last edited by Eva; 23rd February 2010 at 05:05 AM.
  Reply With Quote
Old 23rd February 2010, 04:44 AM   #32
diyAudio Member
 
Join Date: Apr 2003
Location: Southern Tier NY
Default Latest based on Eli and George

Eli, here is my interpretation of what you said. I didn't put any values yet as it is getting close to bedtime.

George, -bias has been increased. If it still is not enough, I can easily series the 5V to the bias winding an gain a few more volts as Eli suggested


Thanks Again
Attached Files
File Type: pdf Visio-Opus5_0_Amp v7.pdf (44.4 KB, 114 views)
File Type: pdf Visio-Opus5_0 Power Supply_v6.pdf (42.2 KB, 83 views)
__________________
Living Life Doing the Waltz in 4/4 meter.
  Reply With Quote
Old 23rd February 2010, 01:50 PM   #33
diyAudio Member
 
Join Date: Apr 2003
Location: Southern Tier NY
Based on Tubelabs comments I drew up this version for peer review and consideration.

I added a 25V rail which at least at this time is regulated with a LM317 set-up for high ripple rejection. I also added a resistor to tie the wiper of the bias pot to the C- should it open up. I am thinking that the impedance of the bias circuit is too high as it stands and may be subject to noise pick-up.

This design does drop the power dissapation of the FET down considerably. (approx 1.2W). The complexity of the power supply increases significantly with the potential of adding more hash into the HT rail due to the addition of another rectification bridge. (It will be schottky)

Bear with my thinking for a moment. In this case if the FET fails it would send the Vg to 25V which would also cause a tube melt down if not caught quick enough. The question I pose is: Is the added complexity worth the benefits? While reducing the demands of the FET, it adds potential other failure modes.

Is there a more eloquent solution in this topology?
Attached Files
File Type: pdf Visio-Opus5_0_Amp v7beta.pdf (44.6 KB, 98 views)
File Type: pdf Visio-Opus5_0 Power Supply_v6beta.pdf (43.4 KB, 69 views)
__________________
Living Life Doing the Waltz in 4/4 meter.
  Reply With Quote
Old 23rd February 2010, 02:40 PM   #34
diyAudio Member
 
tubelab.com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Quote:
I started reading that thread a few days ago and haven't got to the point of you needing to drop the rail. I'll keep reading
I haven't read my own thread in a while, but I believe that we always planned for a pair of rails in the + - 100 to 125 volt range. The method for acheiving this may vary. This is based on my experience with the Tubelab SE amp where the fet runs on the B+ supply. I have had no issues running the Tubelab SE on supplies from 300 to 360 volts or so. A few users have seen fried fets from running their amps in the 400 volt range, and I have set a mosfet on fire and burned up the bias supply by running 480 volts on a board using sweep tubes. There is just too much dissipation in too small of a space.

Quote:
Based on Tubelabs comments I drew up this version for peer review and consideration.......Is there a more eloquent solution in this topology?
I'll tell you what I have been doing. I use a seperate transformer just for cost reasons, but if you are going custom consider this. You need one winding of 160 to 240 volts CENTER TAPPED. Ground the CT and use the typical 4 diode full wave bridge (as found in most SS amps) to generate + - 125 to 150 volts. In my designs I use the Triad N-68X isolation transformer ($11.20 at Mouser). I use the 120 volt "secondary" as the primary, and series the two "primaries" to make a 240 VCT secondary. Filter and / or regulate these supplies as you see fit to arrive at the required voltages. I picked +25 volts for my bench testing because that is the max my bench supply will put out. Many of my test amps have been using + - 150 volts directly from the CLC filtered bridge output.

Top Secret tip!!!! When you make a symmetrical (+- same voltage) supply like this, with almost identical load currents, use a "common mode" choke. This type of choke has two windings, one winding is used in each supply, so that the hum exactly cancels out. You can buy a good one like the One Electron RC-2 ($70 at AES) or you can cheap out and use an isolation (two 120 volt windings) transformer as a choke (this only works where the DC currents are equal).

Quote:
In this case if the FET fails it would send the Vg to 25V which would also cause a tube melt down if not caught quick enough.
+25 will cause a serious current flow through the KT88. +450 or so will cause an instantaneous G1 to cathode arc (proven on a 6L6GC). +150 is in the "I really don't know what will happen" range. I have not had a failure yet. It is a good idea to add a small 125 mA or so fuse in the cathode circuit of each KT88.
__________________
Too much power is almost enough! Turn it up till it explodes - then back up just a little.
  Reply With Quote
Old 23rd February 2010, 03:24 PM   #35
diyAudio Member
 
Join Date: Apr 2003
Location: Southern Tier NY
Default How about this?

George,
Here is a first attempt at your suggestion. I left hash filter in. Never played with common mode choke.

Not even sure how to model it. But I think I caught the basics
Attached Files
File Type: pdf Visio-Opus5_0 Power Supply_v6beta1.pdf (43.4 KB, 86 views)
__________________
Living Life Doing the Waltz in 4/4 meter.
  Reply With Quote
Old 23rd February 2010, 04:18 PM   #36
diyAudio Member
 
boywonder's Avatar
 
Join Date: Jan 2008
Location: So.Cal.
Quote:
Originally Posted by SGregory View Post


boywonder, I am interested in the delay setup.


Here you go. These are both from other threads/members here. I've breadboarded the second one and it works fine (upper schematic). Disregard the lower schema on the second image. I would go with a 45 sec-1 minute amperite time delay relay to allow the heaters to get toasty before applying B+.

When the TDR closes, it energizes the second relay which powers up B+ and takes the TDR out of the circuit to cool for the next power cycle, so even quick power cycles will have a start-up delay.

Of course this adds a little complexity to the PS but the amperites look pretty cool......
Attached Images
File Type: jpg amperite time delay.jpg (38.1 KB, 612 views)
File Type: jpg Amperite time delay 2.JPG (30.1 KB, 610 views)
  Reply With Quote
Old 23rd February 2010, 07:11 PM   #37
diyAudio Member
 
Join Date: Apr 2003
Location: Southern Tier NY
With a common mode choke, Wouldn't the impedance pretty much cancel each other out at the lower frequencies? I asked Electra-print to quote me one and he said a "Common Mode Choke" would only work in the Mhz range. Great for RF but not for ripple.

In my mind I can make it work, but I am confused with his response.

Of course in my mind I can fly without an airplane!!!!!
__________________
Living Life Doing the Waltz in 4/4 meter.

Last edited by SGregory; 23rd February 2010 at 07:13 PM.
  Reply With Quote
Old 23rd February 2010, 08:42 PM   #38
diyAudio Member
 
Join Date: Apr 2003
Location: Southern Tier NY
Output Iron Ordered: Electraprint 4.3K Zp-p Primary @ 6ohm Secondary. 100W 40% UL tap.

Now to tweek PT.
There is alot of good advise here in regard to setting up the rails for the FET. After considering what the FET failure risks are versus the amount of complexity needed to at least mitigate the risk, I am leaning towards letting the tube arc but protecting the OPT. If I was to place a 300mA fuse between the B+ and the OPT, it should acheive that goal. In regard to the FET, I sourced a 900V 2sk2700 equiv (2SK3564) that has 900V Vds capability. It still has a low reverse transfer capacitiance but has 400V more capability than the 2sk3563. Power dissipation aside, this will at least handle start-up voltages.

If FET heat is a problem even with the heat sinks planned, then it should be easy enough to install a small xfmr under the deck to provide 25 or more volts to the drains of the FET. This will minimize the concern of hash on the main core.

What do you think?


boywonder, thanks.
__________________
Living Life Doing the Waltz in 4/4 meter.

Last edited by SGregory; 23rd February 2010 at 08:50 PM.
  Reply With Quote
Old 23rd February 2010, 09:21 PM   #39
diyAudio Member
 
boywonder's Avatar
 
Join Date: Jan 2008
Location: So.Cal.
Default The bias supply has lots of big R's.......

Does using FETs for coupling negate the need to keep the bias resistance below 100K for >35W output according to the KT88 data sheet?
  Reply With Quote
Old 23rd February 2010, 10:36 PM   #40
diyAudio Member
 
tubelab.com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Quote:
Does using FETs for coupling negate the need to keep the bias resistance below 100K for >35W output according to the KT88 data sheet?
The output resistance of the fet is on the order of a few ohms, and the impedance of the fet's power supply better be low, so the requirements are satisfied.

Quote:
With a common mode choke, Wouldn't the impedance pretty much cancel each other out at the lower frequencies? I asked Electra-print to quote me one and he said a "Common Mode Choke" would only work in the Mhz range. Great for RF but not for ripple.
It is not uncommon for someone to resist new ideas. I respect Jacks transformer winding, and I have a pair of his OPT's in my Tubelab SE with 45 tubes. That is the best sounding amp that I have ever built and it has whupped some rather expensive amps when played next to them through Lowthers. Jack told me that PowerDrive, and using silicon to drive big DHT's was stupid. Have you seen some of his new schematics that use a chip amp to drive a DHT through an Electra-Print transformer?

The common mode power supply choke is not even a new idea. In fact I saw it mentioned just a few days ago in a thread on the Tubelab forum. Look at the application notes on good power supply design at the bottom of the page.

5U4G Data sheet. Emission Labs.

Quote:
In my mind I can make it work, but I am confused with his response. Of course in my mind I can fly without an airplane!!!!!
Flying without an airplane is possible, in fact I watched some competitors fly a snowboard just last week!
__________________
Too much power is almost enough! Turn it up till it explodes - then back up just a little.
  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
Modern 3 way crossover design LucasAdamson Multi-Way 2 1st February 2010 10:48 PM
modern allrange speaker Arnold_R Full Range 5 20th October 2009 04:04 PM
Modern cd's - over compression VanHal Music 113 12th March 2008 11:02 PM
Need help- modern replacement for old trannie DeonC Parts 4 7th March 2008 12:52 PM
JLH with modern output transistors diegot Solid State 8 13th August 2006 10:05 PM


New To Site? Need Help?

All times are GMT. The time now is 12:21 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