Choke loaded PSU versus VR tube regulated PSU? - Page 2 - 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 19th March 2008, 04:12 AM   #11
diyAudio Member
 
zigzagflux's Avatar
 
Join Date: Oct 2006
Location: Charlotte, NC
Hmmmmm. Let's back up a little.

If we simply model your first CLC stage, and I have taken the liberty of using a 47uF cap for C2 (which I recommend), and insert a 30mA current source on the output (VR current + load current), your voltage at C2 is only 158V. This will not be nearly enough to get your VR to fire, nor will it allow for any line/load regulation.

I had to increase C1 to 10uF, which brought the voltage at C2 up to about 200V. Now we are in the ballpark for using a 150V regulator tube.

Are you with me so far? If not, we can't really go any further.
  Reply With Quote
Old 19th March 2008, 10:25 AM   #12
G is offline G  United States
diyAudio Member
 
G's Avatar
 
Join Date: Apr 2002
Location: Belleville, IL.
Quote:
Originally posted by zigzagflux
Hmmmmm. Let's back up a little.

If we simply model your first CLC stage, and I have taken the liberty of using a 47uF cap for C2 (which I recommend), and insert a 30mA current source on the output (VR current + load current), your voltage at C2 is only 158V. This will not be nearly enough to get your VR to fire, nor will it allow for any line/load regulation.

I had to increase C1 to 10uF, which brought the voltage at C2 up to about 200V. Now we are in the ballpark for using a 150V regulator tube.

Are you with me so far? If not, we can't really go any further.

Yes. I am with you. I see that I did not put in the values for the capacitors. They are all 100uF capacitors.
__________________
Gavin
  Reply With Quote
Old 19th March 2008, 12:13 PM   #13
JC951t is offline JC951t  Taiwan
diyAudio Member
 
Join Date: Sep 2006
Hi Guys,

You're missing out the point of Chokes in PS.
Putting aside cost, aside from smoothing the ripple,
it does wonders to music reproduction.
Talk to guys who have played around with chokes.
I'm sure they will say the same.
  Reply With Quote
Old 19th March 2008, 12:47 PM   #14
diyAudio Member
 
zigzagflux's Avatar
 
Join Date: Oct 2006
Location: Charlotte, NC
Okay G:

This transformer is a little marginal for the design, but it could be because I do not have the actual parameters modelled correctly in PSUD. I am using 165V at 150 ohm for it. What do you have?

Reason I bring this up is because if I put in your 100u and 100u caps, take a look at V(T1). It is flap topped, indicating the regulation of the transformer has severely limited your output. You obviously want to avoid this condition. You will be faced with this problem regardless if you go with a CCS or not. The CLC is causing it, along with your rated load.

To attempt to remedy it, I have reduced C1 to 3.3 uF, keeping C2 at 100uF. Also reduced your total load to 25mA (10mA load + 15 mA VR). Now, things look much better, though still a little warping of the voltage. Tolerable, IMO, but a cheaper transformer might be making some noise.

With C1 reduced to 3.3uF, I still get 200V at C2, indicating the huge C1 is not really necessary (and in fact makes things worse). Let the choke do its job. Keep C1 small, and maybe higher quality film. Not expensive in this size.

Attached is a tentative design. The 2K resistor will be replaced by your CCS. Note ahead of it, at C2, is 200V, and downstream, at C3, is 150V (0D3 regulated voltage).

I should mention that earlier I stated ripple within 10mV. This is not the ripple present on my regulated B+. There, it is uV. I was simply showing that upstream in the filter section, the ripple I was gettting in PSUD was predicting within 10mV the actual measured ripple. This program is unbelievably accurate, and can be trusted for these simple power supplies. We can get into stability later
Attached Files
File Type: pdf psud.pdf (11.0 KB, 78 views)
  Reply With Quote
Old 19th March 2008, 01:23 PM   #15
G is offline G  United States
diyAudio Member
 
G's Avatar
 
Join Date: Apr 2002
Location: Belleville, IL.
Quote:
Originally posted by zigzagflux
Okay G:

This transformer is a little marginal for the design, but it could be because I do not have the actual parameters modelled correctly in PSUD. I am using 165V at 150 ohm for it. What do you have?

Reason I bring this up is because if I put in your 100u and 100u caps, take a look at V(T1). It is flap topped, indicating the regulation of the transformer has severely limited your output. You obviously want to avoid this condition. You will be faced with this problem regardless if you go with a CCS or not. The CLC is causing it, along with your rated load.

To attempt to remedy it, I have reduced C1 to 3.3 uF, keeping C2 at 100uF. Also reduced your total load to 25mA (10mA load + 15 mA VR). Now, things look much better, though still a little warping of the voltage. Tolerable, IMO, but a cheaper transformer might be making some noise.

With C1 reduced to 3.3uF, I still get 200V at C2, indicating the huge C1 is not really necessary (and in fact makes things worse). Let the choke do its job. Keep C1 small, and maybe higher quality film. Not expensive in this size.

Attached is a tentative design. The 2K resistor will be replaced by your CCS. Note ahead of it, at C2, is 200V, and downstream, at C3, is 150V (0D3 regulated voltage).

I should mention that earlier I stated ripple within 10mV. This is not the ripple present on my regulated B+. There, it is uV. I was simply showing that upstream in the filter section, the ripple I was gettting in PSUD was predicting within 10mV the actual measured ripple. This program is unbelievably accurate, and can be trusted for these simple power supplies. We can get into stability later

I am not following. The transformer is rated at 150mA with both winding in parallel. That is about five times what this circuit will draw, with the OD3, at idle. Below is the PSU I was going to use before I started thinking about regulating it.


Attached Images
File Type: gif clipboard01.gif (17.8 KB, 425 views)
__________________
Gavin
  Reply With Quote
Old 19th March 2008, 02:06 PM   #16
diyAudio Member
 
zigzagflux's Avatar
 
Join Date: Oct 2006
Location: Charlotte, NC
Above you mentioned Hammond 379BX. I don't see that xfmr anywhere on their website. Do you have any data on it?

Regardless, you should see that the peak currents drawn by the first capacitor are well in excess of 150mA when you have the 100uF size installed. That is why the output is sagging. Very typical with capacitor input filters. That is why I recommended lowering the value.

With the design you provided, take a look at V(T1) and I(T1). Very informative.
  Reply With Quote
Old 19th March 2008, 05:43 PM   #17
G is offline G  United States
diyAudio Member
 
G's Avatar
 
Join Date: Apr 2002
Location: Belleville, IL.
Quote:
Originally posted by zigzagflux
Above you mentioned Hammond 379BX. I don't see that xfmr anywhere on their website. Do you have any data on it?

Regardless, you should see that the peak currents drawn by the first capacitor are well in excess of 150mA when you have the 100uF size installed. That is why the output is sagging. Very typical with capacitor input filters. That is why I recommended lowering the value.

With the design you provided, take a look at V(T1) and I(T1). Very informative.

I apologize. The transformer is a 369BX. All I ever look at is the voltage or current supplied to the load. If lowering C1 will help with regulation then by all means proceed.
__________________
Gavin
  Reply With Quote
Old 19th March 2008, 06:15 PM   #18
G is offline G  United States
diyAudio Member
 
G's Avatar
 
Join Date: Apr 2002
Location: Belleville, IL.
Also. I have not purchased the power transformer nor the choke yet so changing the transformer in the model is not a problem. However I am trying to stay with the "X6" sized transformer. You obviously know much more about PSU design than I so by all means feel free to use any 300 series transformer needed. Since I will be using a Ixys bridge you can assume that both windings will be connected in parallel thereby doubling the current rating used for a full wave rectifier. By the way, how did you lower the size of the image of PSUD enough to get the board to accept it?
__________________
Gavin
  Reply With Quote
Old 19th March 2008, 06:27 PM   #19
diyAudio Member
 
zigzagflux's Avatar
 
Join Date: Oct 2006
Location: Charlotte, NC
Very good. That will work. It is actually a center tapped winding, so in PSUD you need to change the rectifier topology to solid state Full Wave, not Bridge. You will get the same result essentially, but only need two diodes instead of four.

I modeled an approximate impedance for the transformer. You would have to measure yours to get more accurate. Close enough for the discussion.

Note that C2 is just under 200V. This is all you have available to fire the VR tube. This is sufficient. Data sheet recommends 185 minimum.

The value for C1, as discussed earlier, affects the voltage at C2. Larger capacitance results in higher C2 voltage, but also higher peak currents through the transformer (which is bad). Values from 3.3 to 4.7 seem to work good, but going as low as 2.2 seems to lower C2 voltage too much for reliable striking. Go for at least 3uF. Pick your favorite capacitor; I like a Solen or CDE 940C; even a 'lytic is fine (*ducks head*). 400V

Again, R1 is the simulated current source. V(R1) is 46V, so your current source will dissipate 46 * 0.025 = 1.15W. No problem; that's a small heatsink, and is cheap for the tight regulation you get. Even under short circuit (disaster) conditions, you only dissipate 5W. A durable design.

Okay, we know the static conditions are fine. VR will take 15 mA, load 10 mA. The tricky part is verifying a successful strike. To do this, you have to look at your load characteristic. In order for the VR to strike, it needs ABOUT 180V, and will vary quite significantly from tube to tube. Therefore, your load voltage will rise from zero to around 180V, without any current being consumed by your VR. Question is, how much current does the LOAD consume at this elevated voltage? If it is equal or greater than 25 mA (all that is available from the CCS), the VR will not strike. If it is only, say 20mA, then the CCS will attempt to drive the full 25 mA, continuing to raise the voltage (up to max 200V) until the VR does strike.

Hope this makes sense. I have designed shunt regulated circuits with resistors that failed to strike for this very reason, so I consider it important to check the startup behavior. Using a CCS makes the calculations easier, of course.

That's about it; I wouldn't add any more components than this. Plenty of threads about CCS design if you need help there.

Two questions, though:
1. Is this a monoblock design, or stereo, where you actually need 2 * 25 mA ?
2. What type of load are we dealing with? Is the consumption essentially constant regardless of music content?
Attached Files
File Type: pdf psud.pdf (16.9 KB, 44 views)
  Reply With Quote
Old 19th March 2008, 06:35 PM   #20
G is offline G  United States
diyAudio Member
 
G's Avatar
 
Join Date: Apr 2002
Location: Belleville, IL.
Quote:
Originally posted by zigzagflux
Very good. That will work. It is actually a center tapped winding, so in PSUD you need to change the rectifier topology to solid state Full Wave, not Bridge. You will get the same result essentially, but only need two diodes instead of four.

I modeled an approximate impedance for the transformer. You would have to measure yours to get more accurate. Close enough for the discussion.

Note that C2 is just under 200V. This is all you have available to fire the VR tube. This is sufficient. Data sheet recommends 185 minimum.

The value for C1, as discussed earlier, affects the voltage at C2. Larger capacitance results in higher C2 voltage, but also higher peak currents through the transformer (which is bad). Values from 3.3 to 4.7 seem to work good, but going as low as 2.2 seems to lower C2 voltage too much for reliable striking. Go for at least 3uF. Pick your favorite capacitor; I like a Solen or CDE 940C; even a 'lytic is fine (*ducks head*). 400V

Again, R1 is the simulated current source. V(R1) is 46V, so your current source will dissipate 46 * 0.025 = 1.15W. No problem; that's a small heatsink, and is cheap for the tight regulation you get. Even under short circuit (disaster) conditions, you only dissipate 5W. A durable design.

Okay, we know the static conditions are fine. VR will take 15 mA, load 10 mA. The tricky part is verifying a successful strike. To do this, you have to look at your load characteristic. In order for the VR to strike, it needs ABOUT 180V, and will vary quite significantly from tube to tube. Therefore, your load voltage will rise from zero to around 180V, without any current being consumed by your VR. Question is, how much current does the LOAD consume at this elevated voltage? If it is equal or greater than 25 mA (all that is available from the CCS), the VR will not strike. If it is only, say 20mA, then the CCS will attempt to drive the full 25 mA, continuing to raise the voltage (up to max 200V) until the VR does strike.

Hope this makes sense. I have designed shunt regulated circuits with resistors that failed to strike for this very reason, so I consider it important to check the startup behavior. Using a CCS makes the calculations easier, of course.

That's about it; I wouldn't add any more components than this. Plenty of threads about CCS design if you need help there.

Two questions, though:
1. Is this a monoblock design, or stereo, where you actually need 2 * 25 mA ?
2. What type of load are we dealing with? Is the consumption essentially constant regardless of music content?
The circuit is essentially a Foreplay preamp using 6SN7s. The current draw will be 5mA per side and since it is a class A there is not going to be much variation in the load.
__________________
Gavin
  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
Trouble choke loading tube in regulated supply phrarod Tubes / Valves 19 14th June 2009 06:13 PM
Choke loaded LTP jane Tubes / Valves 6 8th November 2008 02:25 PM
Choke Loaded ZenV8 RAM Pass Labs 2 30th September 2006 02:01 AM
How to design the straight load line for choke loaded tube ? Antonio Tucci Tubes / Valves 14 30th March 2005 10:05 AM
Choke Loaded Zen hacknet Pass Labs 17 8th April 2004 05:45 PM


New To Site? Need Help?

All times are GMT. The time now is 05:35 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