Beefing up the power supply - 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 15th November 2007, 07:44 PM   #1
diyAudio Member
 
whitelabrat's Avatar
 
Join Date: Dec 2006
Location: Maryland
Default Beefing up the power supply

Howdy ya'll

So I've got a power supply that is 5U4GB rectified and is followed by a 30uF cap. Schematic:

Details

The voltage sags badly so I'd like to stiffen it up. The 5U4GB doesn't like more than 47uF (?) for it's first cap, and I've got a 250uF or a 120uF available. I would figure that a resistor is needed between the 30uF and the 250uF to prevent the 5U4GB from burning up from the inrush current? Would a 80uF motor run do better than the 250uF?

As for the 6X4 side with a choke inline I figure I'm safe replacing the 30uF's with 47uF's.
  Reply With Quote
Old 15th November 2007, 10:45 PM   #2
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
How badly does it sag, and how sure are you that the transformer does not play a large role in this issue? Is that 30uF cap new or original to the amp? If original it is probably bad by now.

You could try a choke and some additional capacitance or a 5AR4. (lower internal impedance.)

What are you using this amplifier for?
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 16th November 2007, 12:44 AM   #3
diyAudio Member
 
Join Date: Apr 2004
Location: Monroe Township, NJ
WLR,

How warm does the power trafo get? Increasing the size of the filter cap. also increasing heating.

I 2nd the 5AR4 idea. If the power trafo is running cool, you can go as high as 60 muF. Also, the smaller forward drop allows you add a choke without worrying about the DCR losses. 1 Henry will isolate the 1st filter cap., rectifier, and trafo from the remainder of the PSU filter. That allows you to safely pile up energy storage in the 2nd position of what will be a CLC filter.
__________________
Eli D.
  Reply With Quote
Old 16th November 2007, 02:30 PM   #4
diyAudio Member
 
whitelabrat's Avatar
 
Join Date: Dec 2006
Location: Maryland
The transformer is cool, and quite large in size. I don't know the current rating, but it doesn't appear to be the problem. Tossing in a 5AR4 will bump up the B+ voltage, probably more than I'd like so I'd like to stick with the 5U4-GB. I'd go with diodes before a 5AR4.

I haven't actually done any measurements to determine the amount of sag, but it can be easily heard by ear. Using my bass guitar there is a sort of compression that happens that isn't tube related. The "attack" doesn't sound right. The attack of a note will require a quick spike of juice an I don't think 30uF can cover it, but does fine during the decay of a note.

Would a CRC filter work or would a resistor pinch the current too much? Noise isn't really a concern for this application. I do agree a CLC filter would be ideal though.
  Reply With Quote
Old 16th November 2007, 03:22 PM   #5
diyAudio Member
 
Tom Bavis's Avatar
 
Join Date: Jun 2002
Location: Macedon NY
What output tubes are you using? With 6L6GCs, there's no need to worry about higher plate voltage - the screen voltage won't change, so the idle current won't go up much. SS would be fine, and you could increase the input cap considerably then. An inrush limiter (CL-70 maybe) on the input AC is always a good idea.

The droop is due to (1) transformer resistance - but you can't do much about that... (2) drop across the rectifier (3) input cap.
  Reply With Quote
Old 16th November 2007, 04:18 PM   #6
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
Quote:
Originally posted by whitelabrat
The transformer is cool, and quite large in size. I don't know the current rating, but it doesn't appear to be the problem. Tossing in a 5AR4 will bump up the B+ voltage, probably more than I'd like so I'd like to stick with the 5U4-GB. I'd go with diodes before a 5AR4.

I haven't actually done any measurements to determine the amount of sag, but it can be easily heard by ear. Using my bass guitar there is a sort of compression that happens that isn't tube related. The "attack" doesn't sound right. The attack of a note will require a quick spike of juice an I don't think 30uF can cover it, but does fine during the decay of a note.

Would a CRC filter work or would a resistor pinch the current too much? Noise isn't really a concern for this application. I do agree a CLC filter would be ideal though.
I am confused by your remark on the 5AR4, a pair of silicon diodes would further raise the supply voltage as compared to any tube rectifier. The 5AR4 typically drops 25 or more volts under load, the 5U4 a bit more at comparable currents and the ss diode <1.2V under load - so your comment makes absolutely no sense. The 5AR4 will be stiffer under load than any 5U4 up to the limits of its ratings. The higher voltage will also help on the attack, just make sure that the output tubes are not over dissipated at the new higher plate voltage and adjust the operating point slightly if they are.

You might want to take a look at some of the interstage coupling time constants as well.

If you haven't replaced the 30uF cap this is the first thing you should do.

The next thing you should do is look at the power supply sag and output stage behavior with a scope, since this is a pa amplifier it is pretty likely you are (close to) saturating the output transformer. Since it is cathode bias you are also driving up the effective output stage bias and the output tubes could be heading towards cutoff on the leading edge of your bass note. (Try a much larger bypass cap on the output stage cathodes.) Also a single pair of 6L6GC aren't a great choice for a bass amplifier. Ever wonder why the AMPEG SVT is rated at 300Wrms? Even modest bass amplifiers are rated at 100W or more in most cases. A better output transformer might be in order with other changes listed below.

Yes beefing up the supply will improve the attack characteristic if the output stage isn't saturating at your normal playing level so this is still a good idea.

I also noticed that this design uses cathode bias - fixed bias would be advisable if you can fit a small transformer. (Or can get the required voltage from the high voltage secondary.) You'll need a raw supply of about -50V - this will improve overload recovery considerably. You could also try something like a high power zener for cathode bias so that the output stage isn't driven into cut off every time you hard clip the output stage. I'd also regulate the screens with zeners.. Fixed bias will also allow you to run fairly cold in class AB and you might get 50W this way.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 16th November 2007, 05:14 PM   #7
diyAudio Member
 
whitelabrat's Avatar
 
Join Date: Dec 2006
Location: Maryland
I've got a LINK that has some good photos and a better schematic of the AU57. Not my photos, but mine appears 100% the same.

I also hope to apply our discussion to other projects, and folks will get some good info from this. Thanks everyone for your input.

Thanks kevinkr. I guess what I mean about the 5AR4 vs 1N4007 diodes is that a lot of things are resolved with diodes if I can take a bump up in B+. I do have a pair of 5AR4's enroute as well. I'll just have to figure out how to rebias the 6L6GC's with the higher voltage.

On a side note I've measured 100Wrms from this guy. That would probably include a heavy dose of distortion too.

Quote:
(Try a much larger bypass cap on the output stage cathodes.)
The 50uF (C-24 from the old schematic)? I have replaced that one with a 330uF. I have no idea what that does, but I'm hoping more is better. I'm slowly replacing all the electrolytics. I have a 33uF to replace the 30uF rated for 500V. And a handful of 47uf's for the preamp stages.

I would do much better with a bigger set of iron, but this guy was only $40 to me. And it is much more fun hot-rodding this oldie!
  Reply With Quote
Old 16th November 2007, 05:57 PM   #8
kevinkr is offline kevinkr  United States
diyAudio Moderator
 
kevinkr's Avatar
 
Join Date: Sep 2004
Location: Boston, Massachusetts
Blog Entries: 6
Hi Whitelabrat,
Yes I would expect at 100Wrms there is a lot of distortion, in theory it should be close to a square wave at this power level..

Tinkering with an old amplifier like this can be a lot of fun, and I now have the sense you are (becoming) aware of the potential limitations present.

Replacing the 5U4 with a 5AR4 should be a drop in, the slight increase in B+ should be tolerable to the output stage as long as you don't overvoltage the electrolytics. (Check with your meter.) Do add a pair of 1N4007 diodes in series with the plates, makes the rectifier set up very robust and will save the rectifier from line transients and arcing while warming up.

Try to figure out how to implement fixed bias in the output stage, I think you'll find that this goes a long way to addressing the attack issue you mentioned. In the meantime that 330uF cap was the right thing to do.

You may find reducing the size of some of the coupling caps will between stages will help. Does your bass have a under saddle piezo pick ups on it in addition to the magnetic pickups? If this is the case some of these produce substantial subsonic crap that will wreak havoc with your amplifier's performance. Again reducing coupling cap values may help and result in a tighter more tuneful bass. (In a previous life I designed a lot of onboard pre-amplifiers for bass guitars with both active and passive mag pickups and active piezo.)

Put a 3 wire (grounded) power cord on this amplifier if you have not done so already. Old amplifiers can be a lethal mix with guitars and other more modern properly grounded things like mixing boards connected to microphones which you can touch while playing..

Most of all keep having fun.
__________________
"Extraordinary claims require extraordinary evidence." - Carl Sagan
  Reply With Quote
Old 16th November 2007, 06:47 PM   #9
diyAudio Member
 
whitelabrat's Avatar
 
Join Date: Dec 2006
Location: Maryland
If it helps to put things into perspective, I sold off a nice SWR solid state amp to fund the purchase of this and other things. I don't gig anymore and I can't very play loud cause it scares the kids. It's primary use will be as a not loud bass amp with my SWR cab, but will also serve as a keyboard amp and guitar amp. It has four inputs with independent preamp stages which I can tune for each application. Kinda nuts, I know.

As for the bass, it's an Alembic Epic with active pickups. The amp does much better with this instrument than it does with the guitar. I'll also agree 6L6's are not ideal for bass amp work, but for modest volume levels they're ok.

I'll look into using an earth ground. I knew a guy who used to get zapped a lot from bad grounding. You could see the spark comming off shoes! Lots of colorful adjectives and verbs followed.

So using something smaller than the 0.047uF's? I would figure the .22 is ok. I'm also working on a SimpleSE that uses a fixed cathode bias. Using some info from Tubelab's website I would guess that a roughly 560 ohm resistor would do. I'll have to measure the voltage to determine the appropriate resistor value.
  Reply With Quote
Old 16th November 2007, 07:23 PM   #10
cerrem is offline cerrem  United States
diyAudio Member
 
Join Date: Jun 2005
Location: San Diego, CA
Sag is a normal part of the circuit....
How much sag is all about the output impeadance of the power supply.... You have a power supply with no closed loop error amp...so it's not high performance, which is OK for the application...
A few things to keep in mind....
The idle voltages are meaningless....and are suppose to be higher than the full power output voltage... Sag is usually accounted for when designing..so the idle voltages will be mucho higher...
Drive the circuit with sig-gen like 100Hz since it's a bass amp...up till you get to the hairy edge just before clipping...
Now, measure the B+ as well as the screen voltage at pin#4....
THIS is the intended design voltages that the original designer worked with when the load-line/plate loading and power output was figured..... Check the RMS power output at this operating point as well... Also check the AC voltage ripple at the B+ and screens to see whats going on with the filter caps....
The best way to linearize the output stage is to regulate the screens...but you need the voltage headroom to do so....
For example... I had an amp with 4x 6550 valves..... The B+ was at 600V at idle and droped to about 520V at full power output....
I set up the screens regulated at 400V ...... It was a nice improvement.....

Chris
  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
LTspice tool for power amp power supply component evaluation andy_c Software Tools 2 23rd August 2009 05:10 PM
question about beefing up my monoblocks sorenj07 Tubes / Valves 0 19th November 2007 03:08 PM
Need advice beefing up my Froys (wmtmw?) MarkB1 Multi-Way 16 10th March 2005 05:20 AM
Beefing up EL34 PP effindi Tubes / Valves 6 17th October 2003 06:38 AM
heater supply (xformer specs are 6.3V 2.5A) as supply for a power LED? jarthel Tubes / Valves 10 21st July 2003 01:30 PM


New To Site? Need Help?

All times are GMT. The time now is 02:26 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