Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: dickjonesify on February 06, 2026, 10:44:52 am
-
I've read a lot of troubleshooting and first time turn on procedure stuff but my problem seems to go all the way back. My voltages all over the amp ('65 Gibson Mercury Medalist) are all too high. Measuring all the way back to the secondary incoming AC, I have 378VAC on either side of the grounded center tap. Measuring DC right off the rectifier diodes, I have 540VDC! (no tubes installed) The schematic says 468V here.
My wall voltage is a bit high at 123VAC but still. Seems much too high. Does that mean the PT is bad? Any insight is appreciated.
My next issue is that the 540VDC goes to every filter cap node(!) while the schematic says it should be stepping down to 465V, 350V and 310V... that's surely related. I'll add that I just put in new Mod filter caps and they seem to be doing their job.
-
With no tubes installed all your B+ voltages will be very high. Plug in the tubes and the voltages will look a lot more reasonable.
-
With no tubes installed all your B+ voltages will be very high. Plug in the tubes and the voltages will look a lot more reasonable.
Ok! Thanks.
I did come across that it would be higher. It just seems WAY higher haha. The amp still has original RCA power tubes so I didn’t want to put them I without checking.
-
Hmm. Still much too high. Having power tubes in brought the voltage at the rectifier down to 510VDC. The next node is about 500VDC.
The phase inverter node and preamp node are both the same at 498-ish. Too too high and it’s odd that they’re the same(?) that’s probably a clue…
-
What idle anode or cathode current are the output valves drawing.
To get realistic HT voltages, the current draw needs to be realistic.
-
Your B+ is only 9% higher than the voltage shown on the schematic. Part of that increased B+ is due to your wall voltage being 5% higher than the schematic. The other 4% difference is likely due to the tube bias being set too cool. Running the bias hotter will bring the B+ down even more. I would modify the bias supply to make it adjustable. And I would build a Buckaroo as seen on page 2 of my amp scrapbook to deal with the higher wall voltage.
https://sluckeyamps.com/misc/Amp_Scrapbook.pdf
-
What idle anode or cathode current are the output valves drawing.
To get realistic HT voltages, the current draw needs to be realistic.
I measured 33mA using this method:
https://www.google.com/search?q=how+do+you+measure+current+on+a+tube&gs_ivs=1#fpstate=ive&vld=cid:62c9e6fe,vid:a_HJKlRLbhM,st:271
Your B+ is only 9% higher than the voltage shown on the schematic. Part of that increased B+ is due to your wall voltage being 5% higher than the schematic. The other 4% difference is likely due to the tube bias being set too cool. Running the bias hotter will bring the B+ down even more. I would modify the bias supply to make it adjustable.
Thank you! It is quite cool at -55VDC.
Any in sight to why the phase inverter node and preamp node are AS high? Quite a bit beyond 10%. They are each downstream running through their own resistors.
And I would build a Buckaroo as seen on page 2 of my amp scrapbook to deal with the higher wall voltage.
https://sluckeyamps.com/misc/Amp_Scrapbook.pdf
Very cool! Thank you. I'll look at this.
-
Any in sight to why the phase inverter node and preamp node are AS high? Quite a bit beyond 10%. They are each downstream running through their own resistors.
Give me some voltage readings. The preamp and phase inverter are all on the 310V node. Maybe you mean the 310v and 330v nodes? Adjusting the value of R58 will affect the 330V and 310V nodes. Adjusting the value of R59 will affect only the 310V node. I would get the 465V node down before tackling the other nodes.
Just using that bucking transformer will give voltages that are perfectly acceptable for this amp.
-
What idle anode or cathode current are the output valves drawing.
To get realistic HT voltages, the current draw needs to be realistic.
I measured 33mA using this method:
https://www.google.com/search?q=how+do+you+measure+current+on+a+tube&gs_ivs=1#fpstate=ive&vld=cid:62c9e6fe,vid:a_HJKlRLbhM,st:271
...
Thanks, unfortunately the link doesn't work for me, just a blank page.
Maybe copy and paste the method?
1R cathode current sensing resistors are a generally preferred method.
My view is that probing an output valve anode is a last resort option, as the results may be inaccurate unless mitigation is taken to avoid oscillation being induced.
-
Give me some voltage readings. The preamp and phase inverter are all on the 310V node. Maybe you mean the 310v and 330v nodes?
I'm the newby so I'll defer to you but looking at the schematic, it shows the 330V node feeding the plates of the phase inverter, no? And 310V feeding both channels' preamp tubes. Maybe my vernacular is wrong haha or I'm misreading (highly likely).
As far as voltages, I've got darn near 500V at all of those points. It's like the resistors are doing nothing(?) I know I'd have to lift one end to read them accurately with a meter but they all measure darn close to the schematic as is. I don't get it.
I would get the 465V node down before tackling the other nodes.
Agreed.
-
Thanks, unfortunately the link doesn't work for me, just a blank page.
Maybe copy and paste the method?
1R cathode current sensing resistors are a generally preferred method.
Whoops! That was the Google search result haha. This is the video...
&t=273s
Yeah, I should get some 1R resistors. Thank you. When I measured each, I got 39mA on one and 81mA on the other so ¯\_(ツ)_/¯ maybe not too accurate.
-
...looking at the schematic, it shows the 330V node feeding the plates of the phase inverter, no?
V3B is the cathodyne phase inverter and it is fed from the 310V source. V4a and v4b are just a driver for the power tubes. V4 is the only tube fed from the 330V source.
-
I should get some 1R resistors.
should be a "best practice" for 99% of amp-builders
-
V3B is the cathodyne phase inverter and it is fed from the 310V source. V4a and v4b are just a driver for the power tubes. V4 is the only tube fed from the 330V source.
Ah! Thank you. I see that now. I’m new to reading schematics and I know what a cathodyne PI is (thanks to Uncle Doug’s PI video) but totally missed it!
-
should be a "best practice" for 99% of amp-builders
What wattage rating should they be?
-
I use 1Ω 1W 1%.
-
I use 1Ω 1W 1%.
Why 1W, isn't 1/2W acceptable in most all amps..?
-
I use 1Ω 1W 1%.
Why 1W, isn't 1/2W acceptable in most all amps..?
Yes. You would need 0.5V at 0.5A to reach 0.25W. Physical size also plays a role. I like the longer and thicker leads of the 1W resistors so I use them also.
-
1W 1R bypassed by a silicon diode makes the cathode current sensing resistor resilient to valve shorts.
https://el34world.com/charts/Schematics/files/Fender/Fender_65_super_reverb.pdf
-
Thanks everyone who's commented so far. I will be replacing the bias caps, adding a trim pot for adjustable bias and adding cathode resistors. I will most likely also build a Buckaroo. Just need to get this order with Amplified Parts finalized.
I spent some times with the amp yesterday. I wish I could tell you what happened haha. Even though I was getting odd results and high voltage readings, I put the rest of the tubes in and connected the speaker cab out of morbid curiosity. At first, I had no sound at all like my previous test. All of the sudden, I had signal! I pulled out the meter and my B+ was still high at 499V or so but my 330V supply was at like 333V and my 310V supply was about 311V. There's also an OA2 regulator tube as part of the trem circuit that was right at its intended 150V. I don't know why they were are right all of the sudden. I haven't been able to make it fail since hahaha.
BUT it's not all sunshine and daisies. It sounds pretty good but does not get loud, well, nearly as loud as a 40 watt, 2 x 6L6 amp should. It also has an unpleasant breakup. I started trying to do the math on the bias and it opened a whole can of worms. More on that to come, probably including a video to more easily explain. Stay tuned.
-
The 1st few minutes of the video you link actually explains most of the questions you have.
What I would do, as well as prior suggestions, is check the value of R60. If it's out of spec, and higher than 7.5k, that could explain some of the extra voltage through the rest of the amp.
-
Riddle me this:
Running through my light bulb current limiter, my voltages are dead nuts on across the board! Without it, they’re still high but not terrible. :dontknow:
Is that normal? I know if I were cranking the amp and playing it that the current limiter would have an affect. I just didn’t realize it would with the amp just turned on.
-
It's normal for the voltage readings to be lower when using a light bulb current limiter. How much AC voltage is on the filaments when using the light bulb current limiter?
Put your light bulb current limiter under the bench. It has served its purpose. Now build a bucking transformer.
-
It's normal for the voltage readings to be lower when using a light bulb current limiter.
Put your light bulb current limiter under the bench. It has served its purpose. Now build a bucking transformer.
Point taken. You are a gentleman and a scholar.
Quick question because I’m more curious than I am electronically educated haha: Could I not just calculate for and then add a voltage drop resistor right before my B+? It would be a higher wattage but wouldn’t need to high of a resistance. Hmm…
-
there are many "work-a-rounds" there's only 1 or 2 solutions
when I miscalculated, I just admitted defeat n bought the proper PT for the job at hand
-
How much AC voltage is on the filaments when using the light bulb current limiter?
How much AC voltage is on the filaments when not using the light bulb current limiter?
-
How much AC voltage is on the filaments when using the light bulb current limiter?
How much AC voltage is on the filaments when not using the light bulb current limiter?
6.7 without
6.6 with
-
After a couple new tubes and more caps, I've got the amp working! I added adjustable bias and got it spot on, 21 watts per tube. It sounds pretty good! Mostly just loud and clean, nothing fancy.
So the amp has two "channels" but they are 99% identical. I decided to pull channel two and do a different circuit's input / tone stack scheme. I removed everything in red, leaving the preamp tubes and their connections to the plates and cathodes. I also pulled that 220K resistor, which is where channel 2 connected to the rest of amp.
My question is: the amp has a decent 60 cycle hum only now that channel 2 is out. Should I ground something? Should I have a resistor simulating the missing volume pot being turned all the way down? Or do the missing channel 2 components have nothing to do with it? But it definitely seemed to.
Either way, it's a short term problem (hopefully). I have parts on the way to replace that channel.
-
if you left V6 hooked up at the cathode and anode, it also needs a ground reference for the grids or the bias voltage can fluctuate widely. doesn't really seem likely that's creating the hum as it's disconnected from the audio circuit, but i'd put 1M's from the grids to ground regardless, or just pull the tube until you replace the channel components. pulling the tube might be instructive as to whether or not it's contributing to the hum.
guessing your earlier issues with poor tone were just related to too cool of a bias on the power tubes. but it is a bit concerning that it went from no output and showing signs of the preamp not really being loaded, then jumped right to more or less proper voltages. i'd look for bad solder joints in the power supply, and probably give the amp a good 'chopsticking', take a solid piece of wood like a chopstick and gently tap components and nudge wires while monitoring preamp voltage nodes, to identify intermittent connections. just make sure the piece you're using is all wood (don't use a pencil), keep one hand in the pocket, and i'd put it back on the dim bulb limiter for that test.
my money's on there being oxidized and intermittent contacts within the power switch, as it is also a DC Standby switch for everything besides the power tubes. your results sound exactly like the primary AC voltage connection was getting made, but not the Standby connection.
many old switches are able to be (carefully) opened up so the contacts can be cleaned, or have a hole where cleaner can be sprayed, i'd highly recommend doing so. but definitely DON'T use deoxit here. if you can't get it open but there's a hole for spray, i'd recommend using CRC QD Electronics Cleaner, then rotate the switch end to end a few dozen times.
if you can get it open in a way you're confident you can reassemble it, the same cleaner works great, but it may be simpler to use 91% or 99% isopropyl alcohol and q tips. if the contacts are really oxidized or pitted, mineral spirits/lighter fluid/naptha may be more helpful, but test a bit on any plastics they come into contact with first to make sure there's no damage. i find a brass brush, especially on the end of a rotary tool, to be amazing at burnishing electrical contacts - tough enough to blast any crap off, but not tough enough to actually score the metal.
bucking transformer is great for reducing primary voltage, but if cost or space is an issue, i find a high resistance NTC thermistor to often be pretty effective at dropping a few volts off the primary. it can be installed anywhere in series with the live or neutral wire, which is handy, but it gets quite hot, so you need to make sure it's at least an inch or two away from wires and other components. to bring the amp back to the 117VAC spec it was designed for, you'll want to drop anywhere from 5-8VAC on the primary... with a 2A fuse, we can assume primary current is about 1-1
5A... which means you'd need the on-resistance of the thermistor to be anywhere from 3-8 ohms, so 5-6 ohms is a good target. none will be designed to have that much resistance at rated current, but you can get that resistance by choosing one rated for much more current than the amp can draw, at the same time giving it a great safety factor. the Vishay MS2212104 fits the bill very well for only $5: https://www.digikey.com/en/products/detail/vishay-ametherm/MS2212104/749846 looking at the datasheet, it's rated for 4A max and would be 0.75 ohms at 4A, or about 2 ohms at 2A, so for your 1-1.5A, that should land around 4-5 ohms. you could also get multiple lower resistance ones and put them in series.
be still my beating heart... is that really a bias circuit with only 145K total grid leak resistance per 6L6?! i can't remember the last time i've seen an amp that even gets that close to the datasheet max rating of 100K! them's some happy control grids, there.
-
if you left V6 hooked up at the cathode and anode, it also needs a ground reference for the grids or the bias voltage can fluctuate widely. doesn't really seem likely that's creating the hum as it's disconnected from the audio circuit, but i'd put 1M's from the grids to ground regardless, or just pull the tube until you replace the channel components. pulling the tube might be instructive as to whether or not it's contributing to the hum.
Funnily enough, pulling V6 made the hum louder(!) But I also have the tremolo disabled cuz the roach was bad. New one incoming. I realized it was still have a decent impact on the sound. After some poking around, I grounded the grid wire into V3A and it quieted a good bit. (I can’t pull that tube because V3B is the PI).
guessing your earlier issues with poor tone were just related to too cool of a bias on the power tubes. but it is a bit concerning that it went from no output and showing signs of the preamp not really being loaded, then jumped right to more or less proper voltages. i'd look for bad solder joints in the power supply, and probably give the amp a good 'chopsticking', take a solid piece of wood like a chopstick and gently tap components and nudge wires while monitoring preamp voltage nodes, to identify intermittent connections. just make sure the piece you're using is all wood (don't use a pencil), keep one hand in the pocket, and i'd put it back on the dim bulb limiter for that test.
Thanks. I have done some poking around. Nothing yet, sadly. The 4K Ω resistor between B+1 and B+2 (R58) is odd. It measures 4K on the dot BUT I have measured zero voltage drop across it multiple times and yet it’s currently doing its job. Can a resistor measure properly and yet still not do its job under the load?
my money's on there being oxidized and intermittent contacts within the power switch, as it is also a DC Standby switch for everything besides the power tubes. your results sound exactly like the primary AC voltage connection was getting made, but not the Standby connection.
many old switches are able to be (carefully) opened up so the contacts can be cleaned, or have a hole where cleaner can be sprayed, i'd highly recommend doing so. but definitely DON'T use deoxit here. if you can't get it open but there's a hole for spray, i'd recommend using CRC QD Electronics Cleaner, then rotate the switch end to end a few dozen times.
if you can get it open in a way you're confident you can reassemble it, the same cleaner works great, but it may be simpler to use 91% or 99% isopropyl alcohol and q tips. if the contacts are really oxidized or pitted, mineral spirits/lighter fluid/naptha may be more helpful, but test a bit on any plastics they come into contact with first to make sure there's no damage. i find a brass brush, especially on the end of a rotary tool, to be amazing at burnishing electrical contacts - tough enough to blast any crap off, but not tough enough to actually score the metal.
If the standby part of the switch was bad, wouldn’t I have no sound?
bucking transformer is great for reducing primary voltage, but if cost or space is an issue, i find a high resistance NTC thermistor to often be pretty effective at dropping a few volts off the primary. it can be installed anywhere in series with the live or neutral wire, which is handy, but it gets quite hot, so you need to make sure it's at least an inch or two away from wires and other components. to bring the amp back to the 117VAC spec it was designed for, you'll want to drop anywhere from 5-8VAC on the primary... with a 2A fuse, we can assume primary current is about 1-1
5A... which means you'd need the on-resistance of the thermistor to be anywhere from 3-8 ohms, so 5-6 ohms is a good target. none will be designed to have that much resistance at rated current, but you can get that resistance by choosing one rated for much more current than the amp can draw, at the same time giving it a great safety factor. the Vishay MS2212104 fits the bill very well for only $5: https://www.digikey.com/en/products/detail/vishay-ametherm/MS2212104/749846 looking at the datasheet, it's rated for 4A max and would be 0.75 ohms at 4A, or about 2 ohms at 2A, so for your 1-1.5A, that should land around 4-5 ohms. you could also get multiple lower resistance ones and put them in series.
be still my beating heart... is that really a bias circuit with only 145K total grid leak resistance per 6L6?! i can't remember the last time i've seen an amp that even gets that close to the datasheet max rating of 100K! them's some happy control grids, there.
I have parts for a Buckaroo arriving today and tomorrow. In the meantime, I did wire in 4 parallel 5W cement resistors after B+. Seems to be doing great.
-
sorry for the brick of text - i get carried away sometimes.
i thought you were only getting no voltage drop across R58, when you were getting no sound, and the 330/310V nodes were too high. are you saying that you've measured no voltage drop across R58, when you're getting output with decent tone, and 330/310v nodes are near expected voltages?
the reason i suspected Standby switch issues, besides the fact that they're common troublemakers, is because you reported at one point having no output and what appeared to be unloaded voltages, which suddenly inexplicably jumped down to the correct values as you got output. this tracks to intermittent contact within the switch. if the switch had lost contact, you would measure no voltage drop across R58.
putting the resistors after B+ should work fine, what values did you use? if you put them in between the rectifier diodes and the first filter cap, they'll create a sag effect for the whole amplifier, similar to a tube rectifier. good luck with the Buckaroo!