Only Edcor power transformer I found with a bias tap and a close enough HV winding is this one.
https://edcorusa.com/collections/tu...-0-400-300ma-50v-bias-12-6v6-3-0-6-3-3a-5v-3a
It's 800V CT 300mA which is more current than I will ever need.
12.6V 3A CT. I can either wire the 12AX7 and 12AU7 as 12 volt heaters, put them across the whole winding and wire each 6L6 across 1/2 the winding which will give me a current of 1.8A or I can just use 1/2 the 12 volt winding which will give me 2.7A
Odds are I will not need to lower the value of the 560 ohm resistor, but if I need to I certainly can, although replacing it with a choke may be the better option.
I have a Magnavox AMP-142 that used dual 5Y3 and has a filter cap then a 1H choke and another filter cap.
Would 1H be fine for this amp or do I need a larger choke value since there's no filter cap at the rectifier?
If so I'll get this one.
https://edcorusa.com/collections/power-chokes/products/xc63-1h-200ma-1h-200ma-choke
If it had the stock power transformer I more than likely would have figured out something else to do for the bias, but it doesn't and has some unknown transformer in it so I'd feel much better using a new power transformer.
Here's the stock power transformer data.
P/N
54-13
Description
Power Transformer (P: 115V; S: 810V@135mA CT, 5V@3A, 6.3V@4A CT)
Used in
A-9, A-9B, A-9C, W-2, W-3, W-3AM, W-3M
That said because the primary is 115V I'd be most likely replacing the stock transformer anyways so it's just as well it was replaced previously.
https://edcorusa.com/collections/tu...-0-400-300ma-50v-bias-12-6v6-3-0-6-3-3a-5v-3a
It's 800V CT 300mA which is more current than I will ever need.
12.6V 3A CT. I can either wire the 12AX7 and 12AU7 as 12 volt heaters, put them across the whole winding and wire each 6L6 across 1/2 the winding which will give me a current of 1.8A or I can just use 1/2 the 12 volt winding which will give me 2.7A
Odds are I will not need to lower the value of the 560 ohm resistor, but if I need to I certainly can, although replacing it with a choke may be the better option.
I have a Magnavox AMP-142 that used dual 5Y3 and has a filter cap then a 1H choke and another filter cap.
Would 1H be fine for this amp or do I need a larger choke value since there's no filter cap at the rectifier?
If so I'll get this one.
https://edcorusa.com/collections/power-chokes/products/xc63-1h-200ma-1h-200ma-choke
If it had the stock power transformer I more than likely would have figured out something else to do for the bias, but it doesn't and has some unknown transformer in it so I'd feel much better using a new power transformer.
Here's the stock power transformer data.
P/N
54-13
Description
Power Transformer (P: 115V; S: 810V@135mA CT, 5V@3A, 6.3V@4A CT)
Used in
A-9, A-9B, A-9C, W-2, W-3, W-3AM, W-3M
That said because the primary is 115V I'd be most likely replacing the stock transformer anyways so it's just as well it was replaced previously.
Last edited:
How would this knob look on this amp?
https://www.tubesandmore.com/products/knob-aluminum-set-screw-skirted-w-line-1125-diameter
The black version.
https://www.tubesandmore.com/products/knob-aluminum-set-screw-skirted-w-line-1125-diameter
The black version.
3 inputs- phono, line, digital.
2 pairs of outs -4 & 8 OT taps wired separately per channel to 2 enclosures-
8- modd'ed Polk T50s
4-Dayton Audio 12" DVC 8 ohm subs in parallel
Unorthodox? perhaps- but effective the sense that the sub/driver coils are isolated from the mid/hi driver coils...
Not so in a typical single feed passive X-over
Jim
Last edited:
Agreed.
Now is it ok to draw 2.7 amps off 1/2 the 12.6 V secondary or should I wire it so the full secondary is used for lower current being drawn.
Now is it ok to draw 2.7 amps off 1/2 the 12.6 V secondary or should I wire it so the full secondary is used for lower current being drawn.
Did some calculating in my spare time today.
For the first 12AX7 stage
10uF cathode cap 2.2k cathode resistor
To get minimal phase shift I will need a 220uF cap.
For the second 12AX7 stage.
To get minimal phase shift I will need a 220uF cap.
Output stage
Now to get zero phase shift I'd need a 2,000uF cap and now we see why fixed bias is better for HI-FI amps that are used full range.
For the first 12AX7 stage
10uF cathode cap 2.2k cathode resistor
To get minimal phase shift I will need a 220uF cap.
For the second 12AX7 stage.
To get minimal phase shift I will need a 220uF cap.
Output stage
Now to get zero phase shift I'd need a 2,000uF cap and now we see why fixed bias is better for HI-FI amps that are used full range.
For the tone control stage.
I'd need a 470uF cap
For those calculations the voltage I put was the voltage on the schematic and I think the calculator was wanting the AC voltage so the currents and power may not be correct. Phase angle, reactance and impedance are correct though.
Coupling cap to first stage (ignoring the phono preamp which has no input coupling cap), second stage, third stage, fourth stage and fifth stage. They all use 1 meg resistors and the tone control circuit is likely close enough to 1 meg impedance to where it will work for it as well.
I'd need a .5uF cap
Coupling cap to output stage.
I chose 1uF as I'm going fixed bias and will need to lower the 470k grid resistors to 220k when going fixed bias, with 1uF giving me basically zero phase shift at 20Hz.
Keep in mind the goal is to eliminate low frequency phase shift which will also eliminate the potential for low frequency instability.
It will also improve the bass response as the amp will then basically be flat to 20Hz provided the output transformer can handle 20Hz properly.
I'd need a 470uF cap
For those calculations the voltage I put was the voltage on the schematic and I think the calculator was wanting the AC voltage so the currents and power may not be correct. Phase angle, reactance and impedance are correct though.
Coupling cap to first stage (ignoring the phono preamp which has no input coupling cap), second stage, third stage, fourth stage and fifth stage. They all use 1 meg resistors and the tone control circuit is likely close enough to 1 meg impedance to where it will work for it as well.
I'd need a .5uF cap
Coupling cap to output stage.
I chose 1uF as I'm going fixed bias and will need to lower the 470k grid resistors to 220k when going fixed bias, with 1uF giving me basically zero phase shift at 20Hz.
Keep in mind the goal is to eliminate low frequency phase shift which will also eliminate the potential for low frequency instability.
It will also improve the bass response as the amp will then basically be flat to 20Hz provided the output transformer can handle 20Hz properly.
The first filter cap is two 40uF in series each having a 100k 1 watt resistor in parallel.
Now I can get up to a 630 volt electrolytic cap.
What voltage should I use?
Now I can get up to a 630 volt electrolytic cap.
What voltage should I use?
In the first graph that says .5uF 1 meg it should be .05uF 1 meg. I'll use the next higher standard value of .68uF.
Also to be on the safe side I'll use 630V electrolytics in the power supply.
If I calculate the peak voltage from the 405Vac HV winding I get 572.67Vdc unloaded. So I at least need 600 volt filter caps.
Also to be on the safe side I'll use 630V electrolytics in the power supply.
If I calculate the peak voltage from the 405Vac HV winding I get 572.67Vdc unloaded. So I at least need 600 volt filter caps.
Last edited:
Now for the first filter cap the final value will be 220uF 600V.
Looking at the amp with its complexity and being somewhat compact, I will not be going 100% stock. The coupling caps and electrolytics will be the value I've upped them to.
Looking at the amp with its complexity and being somewhat compact, I will not be going 100% stock. The coupling caps and electrolytics will be the value I've upped them to.
Two 40uF in series is 20uF. If you are keeping the tube rectifier the you need to respect the max capacitance it can see.
PSUD2 is a good way to see the highest voltage spike to set the max voltage of the caps. Chances are it was correct in the original schematic.
Edit: Looking at the schematic they are probably in series because of the max working voltage. Probably safest to model it in PSUD2.
PSUD2 is a good way to see the highest voltage spike to set the max voltage of the caps. Chances are it was correct in the original schematic.
Edit: Looking at the schematic they are probably in series because of the max working voltage. Probably safest to model it in PSUD2.
The first filter cap is after a 360 ohm resistor so the surge current will be much lower. I may replace the resistor with a 1H choke though.
Last edited:
A better way to understand the issue would be that it's not possible to eliminate low frequency (or high frequency, for that matter) phase shift, but only to move it around. Attempts to "brute force" a solution with all the CR poles pushed down below the OPT primary inductance pole are unstable in practice because the OPT primary inductance varies with signal level, moving its pole with increasing signal in the direction of the CR poles (less stable) and the power supply decoupling time constants. You really want to set a dominant CR pole in a feedback amplifier of multiple poles.
All good fortune,
Chris
The last amp I did this with there was next to no phase shift at 20Hz with the amp staying 100% stable, however that one I used a new Edcor output transformer, whereas this amp uses the original Heathkit output transformer.
Take the last graph I posted.
For the 470k control grid resistor and 1uF cap here's the CR poles.
|p|= 0.33862753849339[Hz]
Take the last graph I posted.
For the 470k control grid resistor and 1uF cap here's the CR poles.
Pole(s)
p = -0.33862753849339[Hz]|p|= 0.33862753849339[Hz]
Not trying to rain on your parade, just reminding us all of basic stability issues. The earliest discussion I'm aware of was in the revision to the DTN Williamson amplifier, way back when, but still applies and quite well explained.
All good fortune,
Chris
All good fortune,
Chris
Agreed.
I think some designs back then were largely dictated by the size and cost of parts, the available technology and whatever marketing determined was the price point they wanted to sell it at.
For instance a first filter cap of 20uF followed by a choke and a second filter cap of 220uF (where the output transformer gets its B+ from) would definitely make the amp sound better for transient peaks and make the amp have better bass response, but the 220uF 600V cap would have been impractical to use from both a size and cost standpoint.
So some performance would be sacrificed to make a reasonably sized affordable amp.
Often I think those who build and restore tube amps use what's always been used for no real reason other than its always been done that way even though for instance in my example a 220uF 600V cap made now is about the same size or smaller than a lot of can caps made back then and very inexpensive.
I tend to test an amp to find its deficiencies and try to fix them within reason.
Using a larger filter cap where the output transformer primary gets its B+ came from a tube amp builders group on Facebook.
I've seen the improvement a larger main filter cap (20,000uF insead of 2,000uF) can do to a solid state amp in that it handles the transient peaks in the audio much better and also handles a steady low frequency sinewave better, but never thought a tube amp could benefit in the same way.
I think some designs back then were largely dictated by the size and cost of parts, the available technology and whatever marketing determined was the price point they wanted to sell it at.
For instance a first filter cap of 20uF followed by a choke and a second filter cap of 220uF (where the output transformer gets its B+ from) would definitely make the amp sound better for transient peaks and make the amp have better bass response, but the 220uF 600V cap would have been impractical to use from both a size and cost standpoint.
So some performance would be sacrificed to make a reasonably sized affordable amp.
Often I think those who build and restore tube amps use what's always been used for no real reason other than its always been done that way even though for instance in my example a 220uF 600V cap made now is about the same size or smaller than a lot of can caps made back then and very inexpensive.
I tend to test an amp to find its deficiencies and try to fix them within reason.
Using a larger filter cap where the output transformer primary gets its B+ came from a tube amp builders group on Facebook.
I've seen the improvement a larger main filter cap (20,000uF insead of 2,000uF) can do to a solid state amp in that it handles the transient peaks in the audio much better and also handles a steady low frequency sinewave better, but never thought a tube amp could benefit in the same way.
My understanding of this area is that it is difficult to look at isolated parts and extrapolate for the whole. For instance, push-pull amps are effective at cancelling some distortion, PSRR I believe it is, so there may not be the necessity to invest in the ultimate smoothing solution.
It's not so much for smoothing as it is to make for a stiffer B+ supply to handle the transient peaks in the audio better.
Also for all but the push pull output tubes, the smaller the filter cap for a stage is made, the more lower frequency audio that will be across the cap. That tends to reduce the bass output somewhat if the filter caps aren't a high enough value.
I have an antique radio from the mid 30's that uses a transformer to drive the push pull tubes. I noticed some audio across the filter cap for the stage that drives the transformer with the audio increasing in amplitude as the audio frequency went lower. Just as a test I put a larger value filter cap at that point and the signal across the cap decreased.
Now I did listen to the Magnavox AMP-142 in stock form before I did the same mods I'm going to do to the Heathkit and it sounded ok, however after the mods, the bass sounded a bit better, because the only thing now reducing the bass at all is the output transformer which is flat to about 25Hz before it starts showing issues reproducing a sinewave at full output power. The mod that helped the most was going fixed bias because it eliminated the cathode bypass cap for the output tubes.
After doing a mod I test the amp to see if said mod actually made an improvement or if it caused an issue.
Also for all but the push pull output tubes, the smaller the filter cap for a stage is made, the more lower frequency audio that will be across the cap. That tends to reduce the bass output somewhat if the filter caps aren't a high enough value.
I have an antique radio from the mid 30's that uses a transformer to drive the push pull tubes. I noticed some audio across the filter cap for the stage that drives the transformer with the audio increasing in amplitude as the audio frequency went lower. Just as a test I put a larger value filter cap at that point and the signal across the cap decreased.
Now I did listen to the Magnavox AMP-142 in stock form before I did the same mods I'm going to do to the Heathkit and it sounded ok, however after the mods, the bass sounded a bit better, because the only thing now reducing the bass at all is the output transformer which is flat to about 25Hz before it starts showing issues reproducing a sinewave at full output power. The mod that helped the most was going fixed bias because it eliminated the cathode bypass cap for the output tubes.
After doing a mod I test the amp to see if said mod actually made an improvement or if it caused an issue.
Last edited: