Now before I replace a single part in the amp I'll replace the replacement power transformer with the Edcor transformer I linked.
It could be that the 5volt difference on the HV winding works better with the 1H choke I'm adding and will get me closer to the stock B+ voltage versus the replacement power transformer with the unknown HV secondary.
I'd almost be willing to bet the transformer has a lower HV secondary voltage which could explain why the 360 ohm resistor was bypassed with a wire.
The stock can caps are held in with nuts and bolts so I'll remove them and put terminal strips there for the new filter caps.
Also before I do anything I need to make sure there hasn't been anything else done to the amp that isn't stock.
It could be that the 5volt difference on the HV winding works better with the 1H choke I'm adding and will get me closer to the stock B+ voltage versus the replacement power transformer with the unknown HV secondary.
I'd almost be willing to bet the transformer has a lower HV secondary voltage which could explain why the 360 ohm resistor was bypassed with a wire.
The stock can caps are held in with nuts and bolts so I'll remove them and put terminal strips there for the new filter caps.
Also before I do anything I need to make sure there hasn't been anything else done to the amp that isn't stock.
Last edited:
So here's what I decided to do.
Given I'm making the rest of the filter caps 33uF 600V I will order an extra for the first filter cap.
I also will order the stock as well as the new value of cathode bypass caps.
The purpose is so I can do a test with the amp close to stock.
I'll then install the upgraded parts and do another test.
That way I have it on paper the improvement if any the upgraded parts caused.
Given I'm making the rest of the filter caps 33uF 600V I will order an extra for the first filter cap.
I also will order the stock as well as the new value of cathode bypass caps.
The purpose is so I can do a test with the amp close to stock.
I'll then install the upgraded parts and do another test.
That way I have it on paper the improvement if any the upgraded parts caused.
The nice thing about using that Edcor transformer is it has enough current on the HV winding to where I can modify the amp should I so desire at some point and get much more than 20 watts out of it.
Ok it looks like I cannot get a 30uF or 33uF 600 volt electrolytic cap from anywhere.
The issue is, there may be a voltage up to 572.67Vdc if the rectifier comes up faster than the tubes or if somehow the tubes are pulled with just the rectifier in place.
I cannot afford to have lower voltage caps blow because of that.
So what can I do?
I'm sort of stuck here.
EDIT:
The only option is either get film caps and hope they fit under the chassis
OR
Find a power transformer that has a lower HV secondary voltage with a bias tap that when used with a 1H choke instead of the 360 ohm resistor will produce 360Vdc with the amp at idle. I could also put a 22uF cap at the rectifier before the choke as that will allow me to use a lower voltage HV secondary than with just a choke.
Another option is to test the transformer that is installed on the chassis first just to see what HV it puts out as it could already be lower than 405Vac which if so will mean I will not need 600V caps.
Honestly I have no clue why Heathkit did it the way they did using a 360 ohm resistor unless they already had a power transformer being produced with that HV secondary voltage and figured they could save money by not having another transformer to keep in stock.
Also if the schematic voltages are correct, the resistor is being ran right at its 7 watt rating given the schematic shows a 50 volt drop across the resistor and ohms law shows 50 volts across 360 ohms is 7 watts.
The issue is, there may be a voltage up to 572.67Vdc if the rectifier comes up faster than the tubes or if somehow the tubes are pulled with just the rectifier in place.
I cannot afford to have lower voltage caps blow because of that.
So what can I do?
I'm sort of stuck here.
EDIT:
The only option is either get film caps and hope they fit under the chassis
OR
Find a power transformer that has a lower HV secondary voltage with a bias tap that when used with a 1H choke instead of the 360 ohm resistor will produce 360Vdc with the amp at idle. I could also put a 22uF cap at the rectifier before the choke as that will allow me to use a lower voltage HV secondary than with just a choke.
Another option is to test the transformer that is installed on the chassis first just to see what HV it puts out as it could already be lower than 405Vac which if so will mean I will not need 600V caps.
Honestly I have no clue why Heathkit did it the way they did using a 360 ohm resistor unless they already had a power transformer being produced with that HV secondary voltage and figured they could save money by not having another transformer to keep in stock.
Also if the schematic voltages are correct, the resistor is being ran right at its 7 watt rating given the schematic shows a 50 volt drop across the resistor and ohms law shows 50 volts across 360 ohms is 7 watts.
Last edited:
Here's what I don't understand.
Why would any company build something that could have a fault easily done by the user that will blow all the filter caps?
Why would any company build something that could have a fault easily done by the user that will blow all the filter caps?
Ok so here's where I am now.
I can easily get 33uF 500V and 220uF 500V caps so that means I can use a transformer with up to a HV of 340-0-340V if I use a filter cap right after the rectifier.
Once the amp is restored I'll power the rectifier from a 5Vac transformer and the tube heaters from a 6.3Vac transformer. That way I can vary the AC input to the power transformer and see what HV I need to get 360Vdc B+ at the first filter cap. I'll try it with a CLC filter and then a LC filter.
I'll possibly use this choke.
https://edcorusa.com/collections/po...QIZ4nyK277iqOWL7nGysDP4B-83AekslC0KV_hv4E6a88
Doing those tests make it easier to find a power transformer as I know exactly what voltage range I can use for the HV winding.
I can easily get 33uF 500V and 220uF 500V caps so that means I can use a transformer with up to a HV of 340-0-340V if I use a filter cap right after the rectifier.
Once the amp is restored I'll power the rectifier from a 5Vac transformer and the tube heaters from a 6.3Vac transformer. That way I can vary the AC input to the power transformer and see what HV I need to get 360Vdc B+ at the first filter cap. I'll try it with a CLC filter and then a LC filter.
I'll possibly use this choke.
https://edcorusa.com/collections/po...QIZ4nyK277iqOWL7nGysDP4B-83AekslC0KV_hv4E6a88
Doing those tests make it easier to find a power transformer as I know exactly what voltage range I can use for the HV winding.
Using two 68uF capacitors in series, say 350VDC, with balancing resistors (which double as bleeder resistors) will work at that B+, and is equivalent to 34uF.
Perhaps, however I've never really liked caps in series like that.
First I gotta see what the unloaded B+ will be then I can decide where to go from there based on what power transformer I can get.
First I gotta see what the unloaded B+ will be then I can decide where to go from there based on what power transformer I can get.
Looks like it will be either Christmas if I get any Christmas money or February when I do my tax returns before I restore this amp.
While looking for something else last night I stumbled upon a Hammond 272FX power transformer I had forgotten I had.
https://www.hammfg.com/part/272FX?referer=1052
115 volt primary
600 V CT at 172mA
6.3 V CT at 5A
5 V CT at 3A
Running on 120Vac the voltage will be
626 Vrms
6.57 Vrms
5.21 Vrms
Those should be acceptable voltages.
Will that be a suitable transformer or will the B+ be too low?
It also does not have a bias tap either.
https://www.hammfg.com/part/272FX?referer=1052
115 volt primary
600 V CT at 172mA
6.3 V CT at 5A
5 V CT at 3A
Running on 120Vac the voltage will be
626 Vrms
6.57 Vrms
5.21 Vrms
Those should be acceptable voltages.
Will that be a suitable transformer or will the B+ be too low?
It also does not have a bias tap either.
Would that Hammond transformer work for this amp?
Not going to be able to use the amp for the original purpose and will be using it for more of a LO-FI application where I do not need the mods I had planned to make it a better HI-FI amp so I'll likely restore it to stock.
It's potentially going to be used with an antique radio to drive a larger speaker to reproduce the lower audio frequencies (with a -3dB down point of about 40 Hz) the radio's speaker cannot reproduce as I like to use it to listen to big band and swing music. So the slight roll off of the bass the amp in its current configuration has will be fine.
Just need to figure the power supply out.
If the B+ is too high maybe I can use a field coil speaker to drop some voltage and also provide better filtering of the B+ supply.
Not going to be able to use the amp for the original purpose and will be using it for more of a LO-FI application where I do not need the mods I had planned to make it a better HI-FI amp so I'll likely restore it to stock.
It's potentially going to be used with an antique radio to drive a larger speaker to reproduce the lower audio frequencies (with a -3dB down point of about 40 Hz) the radio's speaker cannot reproduce as I like to use it to listen to big band and swing music. So the slight roll off of the bass the amp in its current configuration has will be fine.
Just need to figure the power supply out.
If the B+ is too high maybe I can use a field coil speaker to drop some voltage and also provide better filtering of the B+ supply.
Last edited:
It will help to know what modifications you have in mind at this point, after many months, for the “more LO-FI” version? Are you still planning control grid bias? It should be quite possible to make a LO-FI amp with the A9 parts and 272FX.
I suggest you get PSUD2 and simulate your power supply options. If you are not already using it get it here: https://www.duncanamps.com/psud2/
I suggest you get PSUD2 and simulate your power supply options. If you are not already using it get it here: https://www.duncanamps.com/psud2/
For the LO-FI version I plan on leaving the amp in its stock configuration only replacing the resistors and caps with stock values.
The only liberties I may take is with the power supply given I do not have the original power transformer.
The unloaded DC voltage would be 442.5Vdc. Not sure what the loaded voltage will be. Long as I get close to 360Vdc on the 6L6 plates then I'll know the supply is right. Depending on the final B+ voltage I might be able to power a field coil which will provide additional power supply filtering.
Also given the unloaded B+ voltage is under 450Vdc I should be able to use commonly available 450Vdc caps for the filter caps.
I forget exactly where I may have posted it, but I'm thinking I did measure the voltages of the power transformer that is on the chassis, but I can measure them again when I get home today. Long as the HV winding is close to 305-0-305Vdc then I might just use it.
The only liberties I may take is with the power supply given I do not have the original power transformer.
The unloaded DC voltage would be 442.5Vdc. Not sure what the loaded voltage will be. Long as I get close to 360Vdc on the 6L6 plates then I'll know the supply is right. Depending on the final B+ voltage I might be able to power a field coil which will provide additional power supply filtering.
Also given the unloaded B+ voltage is under 450Vdc I should be able to use commonly available 450Vdc caps for the filter caps.
I forget exactly where I may have posted it, but I'm thinking I did measure the voltages of the power transformer that is on the chassis, but I can measure them again when I get home today. Long as the HV winding is close to 305-0-305Vdc then I might just use it.
Last edited:
With 600 VCT and SS rectification you should have no difficulty dialing in the 360 Vdc for the original B+. That large R48 wasted a lot of heat. Try to avoid or reduce that.
Download PSUD2 and play around with the PS component values. It is not difficult to do. Good luck.
Download PSUD2 and play around with the PS component values. It is not difficult to do. Good luck.
If I do need to drop voltage I'll use a speaker with a field coil as that will also provide better filtering for the power supply. I'd need to figure the voltage drop necessary and what resistor gets me close then find a speaker with a field coil close to the required resistance. If none are readily available I can parallel a resistor with the field coil to make the required resistance.
Measured the power transformer and the HV winding is 300-0-300Vdc unloaded.
So I'll try that one first.
The unloaded B+ will be 424.2Vdc if I put a filter cap at the 5V4 cathode. Will 450V caps be a high enough working voltage?
Also for the volume pot I have a 1 meg pot, but I also have a 100k pot I would prefer to use. Will using the 100k pot cause any issues provided I up the .05uF coupling cap to .5uF? The preceding stage uses a 100k plate resistor.
So I'll try that one first.
The unloaded B+ will be 424.2Vdc if I put a filter cap at the 5V4 cathode. Will 450V caps be a high enough working voltage?
Also for the volume pot I have a 1 meg pot, but I also have a 100k pot I would prefer to use. Will using the 100k pot cause any issues provided I up the .05uF coupling cap to .5uF? The preceding stage uses a 100k plate resistor.
Last edited:
What I'm tempted to do is replace R9 the 18k resistor with a 1k resistor and a 20k trimpot. That way I can adjust the input signal so that the 12AX7 stage doesn't distort given the input signal will likely be several volts.
For a 10 volt input with the resistor being 18k I get 177mV out. For 1K I get 10mV. For 21k I get 206mV. That's the calculated voltages.
What I will first do is restore the amp stock then feed a signal into the tuner in and see at what input voltage the 12AX7 stage starts to distort. I can then remove the driver tube in the radio and see how much signal it takes to cause the first audio stage to distort. I can then compare the two numbers and see how close they are to each other. If the radio puts out more signal than what will distort the 12AX7 stage I can either reduce the value of the 18k resistor to lower the signal level (a voltage divider calculator will tell me what value it needs to be) or remove C5 which lowers the gain of the 12AX7 stage.
For the power supply I will rebuild it without the series resistor then power the amp to see what the B+ is. If it's over 360Vdc I'll see if I have a choke that I can put after the rectifier and first filter cap which will provide better filtering and drop the B+ some. To provide even better filtering I'll use an LC resonant circuit calculator to calculate the cap value I need for 120Hz. I did that with a power supply that due to its compact size I couldn't fit larger filter caps inside and it cut the ripple voltage from around 200mV to 12mV. So if I use a 10H choke I'd need a .175uF cap which rounded up will be .18uF a commonly available value. If I do use a choke I'll likely mount it where the can caps are given the new filter caps will go under the chassis.
Once the amp is rebuilt and tested I then need to figure out what to do for a speaker and speaker cabinet.
One question about the feedback.
Does it connect to the impedance tap being used or does it stay connected to a certain tap?
For a 10 volt input with the resistor being 18k I get 177mV out. For 1K I get 10mV. For 21k I get 206mV. That's the calculated voltages.
What I will first do is restore the amp stock then feed a signal into the tuner in and see at what input voltage the 12AX7 stage starts to distort. I can then remove the driver tube in the radio and see how much signal it takes to cause the first audio stage to distort. I can then compare the two numbers and see how close they are to each other. If the radio puts out more signal than what will distort the 12AX7 stage I can either reduce the value of the 18k resistor to lower the signal level (a voltage divider calculator will tell me what value it needs to be) or remove C5 which lowers the gain of the 12AX7 stage.
For the power supply I will rebuild it without the series resistor then power the amp to see what the B+ is. If it's over 360Vdc I'll see if I have a choke that I can put after the rectifier and first filter cap which will provide better filtering and drop the B+ some. To provide even better filtering I'll use an LC resonant circuit calculator to calculate the cap value I need for 120Hz. I did that with a power supply that due to its compact size I couldn't fit larger filter caps inside and it cut the ripple voltage from around 200mV to 12mV. So if I use a 10H choke I'd need a .175uF cap which rounded up will be .18uF a commonly available value. If I do use a choke I'll likely mount it where the can caps are given the new filter caps will go under the chassis.
Once the amp is rebuilt and tested I then need to figure out what to do for a speaker and speaker cabinet.
One question about the feedback.
Does it connect to the impedance tap being used or does it stay connected to a certain tap?
Last edited:
One change I will make is using a 5AR4 instead of a 5V4 given the 5AR4 has a lower voltage drop at 17V for a current of 225mA versus the 5V4 which is 25V at 175mA. Same heater current though.
I would think so. Check it during startup when you have your final PS assembled, to see how high it actually goes. 450 V caps should do it. Most caps can tolerate a momentary overvoltage for this purpose anyway. Check the specs on the caps you use before replacing them with higher voltage ones.
I’m not sure on this one. You could try to determine if the output impedance on the 12ax7 under its operating conditions is sufficiently low to drive a 100k pot. Or try it with .5 coupling caps, and switch if you don’t like what you measure or hear.
Regarding R9 - I don’t understand your concern with matching a tuner with input of several volts that will overdrive the amp input. Most tuners I know have output pots to control that.
You plans with the power supply and 5AR4 sounds good to me. Good luck.
The input on the tuner input will be a Philco 118 radio with the output being taken from the first audio stage plate through a low pass filter. There will be several volts of audio at that point.
I will definitely try the 100k pot first and measure the voltage across it with an input of 400Hz and will then repeat the test with the 1 meg pot.
I will definitely try the 100k pot first and measure the voltage across it with an input of 400Hz and will then repeat the test with the 1 meg pot.
Last edited: