Heater Hum

[dankinzelman]

Hi there,

I am working to make some improvements to a series of FBT 500R amps. After fixing some power supply issues (bad caps, PI powered off the wrong node), the amps are much quieter, but the highest gain circuits (channel 2 and reverb circuit) still have some 50hz hum when turned up. Based on my reading, this is likely hum from the AC heaters feeding all tubes. The heater winding has a grounded center tap, and wiring does not follow best practices, ie heater wires are not twisted together, and they run close to all tube sockets in the preamp section (see attached photo, heater wires are the brown wires I'm pointing to with my chopstick, you can see they are sort of wrapped around each socket, albeit with a bit of standoff distance).

My thinking is I can 1. elevate heater center tap or 2. rewire all preamp heaters using twisted wires or 3. do both.

I think elevating the heater center tap would be easier and I tried designing a circuit to do it (see attachment). Will this work, and if so, is there danger of the heater AC creeping back into the PS node I've chosen?

Would I be better off rewiring the whole heater circuit? Advice?

[Merlin]

I'd try a humdinger. You can just tack it in to start with (disconnect the CT), to see if it buys you any improvement or not.

[Latole]

I often make this mods with two 100 ohms , work very well.

[Willabe]

I often make this mods with two 100 ohms , work very well.

The PT has a CT for the heater wind.

My thinking is I can 1. elevate heater center tap or.....

 

[Willabe]

......My thinking is I can 1. elevate heater center tap or 2. rewire all preamp heaters using twisted wires or 3. do both.

I'd try a humdinger. You can just tack it in to start with (disconnect the CT), to see if it buys you any improvement or not.

I like doing both, elevate the heaters with a hum dinger pot. If you have the room, it's not many parts, doesn't cost much to do.

But you probably will have to drill a hole in that chassis. 

Here's my Princeton 5F2a clone with hum dinger pot and elevated heaters. Schematic shows dc stand off/elevated voltage circuit on the bottom left, and hum dinger pot, bottom right.

[Latole]

I often make this mods with two 100 ohms , work very well.

The PT has a CT for the heater wind.

Did it work , is it connected in the secondary  ?

[pdf64]

...
I like doing both, elevate the heaters with a hum dinger pot. 
 ...

If the heater circuit is DC elevated, I wonder if a humdinger can / does have any benefit compared to balancing the heater circuit with a pair of equal, low value resistors?
Did you test for that?

[Latole]

...
I like doing both, elevate the heaters with a hum dinger pot. 
 ...

If the heater circuit is DC elevated, I wonder if a humdinger can / does have any benefit compared to balancing the heater circuit with a pair of equal, low value resistors?
Did you test for that?

You are right . I never use DC on heater and all the amp I fix are dead quiet

[Willabe]

If the heater circuit is DC elevated, I wonder if a humdinger can / does have any benefit compared to balancing the heater circuit with a pair of equal, low value resistors? Did you test for that?

No, never tested for any difference.

I got the elevated dc stand off with hum dinger pot from Kevin O'Conner TUT books.

On a new build it's easy to put in and only costs you a pot. 

[labb]

Try the humdinger as Merlin suggested. Easy to do. You should be alright with the parallel heater wiring.

[trobbins]

The humdinger pot may help with capacitive coupling from heater voltage over to input stage grids, as it can help neutralise the effect of such coupling (but could be worthwhile doing a temporary check by powering the heater wires going to the preamp circuitry with a 6V battery to see if that is a major hum contributor).  In contrast, the elevated dc voltage for the heater may help suppress any ac leakage through the heater-cathode interface and into the input stage cathode circuit (but worthwhile also tube-rolling to see if that can help lower the hum, as not all tubes have the same leakage).

[pdf64]

The humdinger pot may help with capacitive coupling from heater voltage over to input stage grids, as it can help neutralise the effect of such coupling ...

I'm stuck trying to see how a humdinger could be more effective at helping with that than a pair of equal and suitably low value resistors

[trobbins]

Parasitic capacitance from each heater wire/side may not be balanced, such that some leakage occurs - with more leakage into a grid from one heater side than the other - that can be re-balanced by the humdinger pot.  However the heater voltage is not a pure sinewave, so some leakage frequencies may still leak into the grid, especially if the mains has significant distortion or flat-topping, and if the rectifier's operation is distorting the heater voltage (eg. likely worse with ss rectifier).

In contrast, a heater ct or fixed humdinger may still allow some unbalanced leakage (eg. checked by using a temporary 6V battery to feed the wiring into the preamp section).

[dankinzelman]

Wow, this is an active thread! Thanks everyone.

I tried disconnecting the CT and using the humdinger (~600 ohm pot). Lowest hum was with one side of the heater circuit grounded (ie the pot all the way to one side). The pot all the way on the other side resulted in greatest hum. I need to find a fast way to swap between the artificial tap humdinger and the stock (grounded CT) to better compare.

What kind of battery is 6V?

[SEL49]

Look here...

https://www.walmart.com/ip/Eveready-Super-Heavy-Duty-6V-Spring-Terminal-Carbon-Zinc-Lantern-Battery-1209/10242712?classType=REGULAR&from=/search

[trobbins]

Another alternative is a 6V 4Ah SLA lead-acid, but yes anything that can support one or two input stage heaters (eg. 0.3 or 0.6A) for perhaps a minute or two, is the kind of quick test that may assist.  Perhaps 4x 1.5V D cells, or C-cells.

[Merlin]

I tried disconnecting the CT and using the humdinger (~600 ohm pot). Lowest hum was with one side of the heater circuit grounded (ie the pot all the way to one side).

That suggests it would indeed be beneficial for you to change the heater wiring to a proper twisted arrangement.

[dankinzelman]

Ok, what size wire do I need for this? I can just redo the offending stages, right? Is it worth trying to elevate the heaters first?

[Merlin]

Ok, what size wire do I need for this? I can just redo the offending stages, right?

If you just rewire a few preamp valves you can use whatever wire you like. 7/0.2 stranded?

Is it worth trying to elevate the heaters first?

Sure, try it if you like. It's easier to try a small mod like that first, than to redo wiring.

[acheld]

The wiring as shown in your photo appears neat and pretty well tucked away.   I'm not saying it might not be improved, but I doubt that wire dress is the problem here. (See Sluckey's famous fotos  https://www.sluckeyamps.com/dual_lite/dual_lite.htm).

I am intrigued by the statement that the hum goes nearly goes away when the balance pot grounds one side of the heater circuit.  That's not right.   Is there a fault in the heater windings? 

[Willabe]

I can just redo the offending stages, right?

How can you tell which is an offending stage?

You can't. If you redo the heater wiring you do the whole heater string.

[Willabe]

I tried disconnecting the CT and using the humdinger (~600 ohm pot). Lowest hum was with one side of the heater circuit grounded (ie the pot all the way to one side).

That suggests it would indeed be beneficial for you to change the heater wiring to a proper twisted arrangement.

Can you explain further for us what is happening?

[Merlin]

Can you explain further for us what is happening?

With twisted heater wiring the hum fields are (nearly) symmetrical. Induced hum should then be mainly sneaking through roughly symmetrical pathways into the audio circuit, so we expect the point of minimum hum to be roughly central on a humdinger.

But with non-twisted wiring like this, those hum pathways are hard to predict and may be non-symmetrical, so it's no unsual to find minimum hum occurs with the humdinger way over to one side. It's not guaranteed, but the humdinger is hinting that the lack of twisting is causing the bulk of the hum.

But yeah, re-doing it is a pain, so try elevating / DC / tube swapping first. You can't solve a problem until you know what the problem is.

[Willabe]

Thank you Merlin.

[acheld]

the humdinger is hinting that the lack of twisting is causing the bulk of the hum.

The humdinger response is really only hinting that there is an asymmetry in heater circuit. A cold solder joint, a fault in the (secondary) heater windings, or even a bad tube can cause these symptoms.

While I do twist my heaters (old dog, old habits), I have become convinced that twisting is over-rated as a solution to heater hum. 

And in this amp, we don't really know its' full history.   When did the hum develop? 

[labb]

A couple of things before redoing the heater wiring. Voltage  at the transformer each side to ground. Resistance each side to ground. As to the parallel heater wiring: the last three amps I have built I use the parallel wiring. Only difference is I went over head instead of tucked in the corner. One of the amps was high gain Marshall type. They do not hum. Just saying. Last thing I would do would be to redo the wiring. Pain in the***

[dankinzelman]

Regarding the offending stage, there is no hum whatsoever on channel one, even with volume cranked. Channel two introduces hum when I turn the volume up, and adding reverb increased it.

I tried elevating the heaters and it didn't seem to make any difference so I redid the wiring on the reverb and channel 2 gain stages, flying twisted pair arching from the tremolo tube over to the channel 2 gain and reverb tubes wired in parallel. It may have improved somewhat, it seems like at least the reverb return is humming less. The hum is still there though on channel two (even with reverb off). It's not extreme, but it's annoying from about 50% volume pot travel up, and would be nice to get rid of it.

I will try swapping tubes again, lots of stuff changing so I have a lot of variables here.

[Latole]

The wiring as shown in your photo appears neat and pretty well tucked away.   I'm not saying it might not be improved, but I doubt that wire dress is the problem here. (See Sluckey's famous fotos  https://www.sluckeyamps.com/dual_lite/dual_lite.htm).

I am intrigued by the statement that the hum goes nearly goes away when the balance pot grounds one side of the heater circuit.  That's not right.   Is there a fault in the heater windings?

I agree !

[Willabe]

Regarding the offending stage, there is no hum whatsoever on channel one, even with volume cranked. Channel two introduces hum when I turn the volume up, and adding reverb increased it.

If it's really hum in the heater wiring, then the hum is in the WHOLE heater string, not just at 1 or 2 tubes or just 1 channel. All the tubes are powered from the same heater string. So if the heater string is causing channel 2 to hum, then it has to also cause channel 1 to hum too. 

I tried elevating the heaters and it didn't seem to make any difference so I redid the wiring on the reverb and channel 2 gain stages, flying twisted pair arching from the tremolo tube over to the channel 2 gain and reverb tubes wired in parallel. It may have improved somewhat, it seems like at least the reverb return is humming less. The hum is still there though on channel two (even with reverb off). It's not extreme, but it's annoying from about 50% volume pot travel up, and would be nice to get rid of it.

What this seems like is when you ran those stages heater wires above, it got the heater wires away from those tubes input grid wires. So it's getting injected into the grids.

That's why elevating the heaters didn't help. 

Looking at the chassis picture you posted it looks like some of the wires going to the preamp tube sockets they ran the wires parallel. It's a short distance, but it might be enough to cause some hum. When wires cross at a 90 degree angle the least amount of cross talk contamination occurs.   

[dankinzelman]

Regarding the offending stage, there is no hum whatsoever on channel one, even with volume cranked. Channel two introduces hum when I turn the volume up, and adding reverb increased it.

If it's really hum in the heater wiring, then the hum is in the WHOLE heater string, not just at 1 or 2 tubes or just 1 channel. All the tubes are powered from the same heater string. So if the heater string is causing channel 2 to hum, then it has to also cause channel 1 to hum too. 

...

Looking at the chassis picture you posted it looks like some of the wires going to the preamp tube sockets they ran the wires parallel. It's a short distance, but it might be enough to cause some hum. When wires cross at a 90 degree angle the least amount of cross talk contamination occurs.

Understood the hum is in the whole heater circuit, but it only seems to be reaching the speaker by passing through the gain stage of channel 2 - I presume this is because the gain in channel 2 is higher, but I will try swapping channel 1 and 2 tubes and see if that fixes it. I will also try seeing if I can move some grid wires around to avoid the heater wiring.

[Willabe]

Understood the hum is in the whole heater circuit, but it only seems to be reaching the speaker by passing through the gain stage of channel 2 - I presume this is because the gain in channel 2 is higher, but I will try swapping channel 1 and 2 tubes and see if that fixes it. I will also try seeing if I can move some grid wires around to avoid the heater wiring.

No.

You haven't swapped the preamp tubes with known good, non humming tubes yet? 

[Willabe]

I never use DC on heater and all the amp I fix are dead quiet.

It allows you to use tubes the have leakage from the factory. Has nothing to do with lead dress.

The humdinger pot may help with capacitive coupling from heater voltage over to input stage grids, as it can help neutralise the effect of such coupling (but could be worthwhile doing a temporary check by powering the heater wires going to the preamp circuitry with a 6V battery to see if that is a major hum contributor).  In contrast, the elevated dc voltage for the heater may help suppress any ac leakage through the heater-cathode interface and into the input stage cathode circuit (but worthwhile also tube-rolling to see if that can help lower the hum, as not all tubes have the same leakage).

 

[Willabe]

Understood the hum is in the whole heater circuit, but it only seems to be reaching the speaker by passing through the gain stage of channel 2 - I presume this is because the gain in channel 2 is higher,

I just saw the schematic you posted for this amp in another thread. Both channels are the same except for a coupling cap value difference.

Please post that schematic here and the schematic for the reverb.

[HotBluePlates]

I am working to make some improvements to a series of FBT 500R amps. ...

... The hum is still there though on channel two (even with reverb off). ...

Kill one Monster at a time.  So for now, leave the Reverb off.

Do you have any capacitors ≥25µF?  Have you already replaced the cathode bypass caps of the first 2 gain stages of Channel 2?

   - One type of hum can happen when the cathode is not solidly bypassed to ground.  Any cap that 20+ microfarads can be tacked in across the cathode resistors of the first gain stages in Channel 2 (when the Reverb is off, these are the only bits unique to Channel 2).

   - If the hum goes away, you could possibly kill the hum with a replacement tube.  However, the new tube could also have the same defect that gives hum due to heater-to-cathode leakage.

   - However, if you don't need/want a small cathode bypass capacitance for some tonal reason (like boosting mids/treble relative to lows), then replacement cathode bypass caps can kill the hum.  And when you use crazy-big caps (like the tweed Bassman's 250µF), you kill leakage hum even when the tube is defective (though the low end might sound boomy).



For a test, you don't even need to remove the existing bypass caps; just tack-solder the new caps across the old ones.  You'll know quickly whether this resolves the hum issue.

[dankinzelman]

I just saw the schematic you posted for this amp in another thread. Both channels are the same except for a coupling cap value difference.

Please post that schematic here and the schematic for the reverb.

Schematic is below.

Kill one Monster at a time.  So for now, leave the Reverb off.

Do you have any capacitors ≥25µF?  Have you already replaced the cathode bypass caps of the first 2 gain stages of Channel 2?

All cathode bypass caps have been replaced with 22uf 50V caps.

Also I just realized the ch. 2 gain and reverb tubes are on the side of the chassis closest to the mains transformer (preamp is in a separate metal housing from PI and power tubes). So I'll also try taking the amp out of the chassis and seeing what happens if I put some distance between them.

Understood the hum is in the whole heater circuit, but it only seems to be reaching the speaker by passing through the gain stage of channel 2 - I presume this is because the gain in channel 2 is higher, but I will try swapping channel 1 and 2 tubes and see if that fixes it. I will also try seeing if I can move some grid wires around to avoid the heater wiring.

No.

You haven't swapped the preamp tubes with known good, non humming tubes yet? 

Can you please explain 'No'? I'm not doubting you, just trying to understand your reasoning.

I have three of these amps. All three have had all electrolytics (including cathode bypass caps) replaced. All three have a different assortment of tubes, including one with all new tubes. The only constant across all three amps is they still have similar hum (but only in channel 2, channel 1 is very quiet even at max volume).

I am traveling for work now, but when I get back home I will try swapping ch. 1 and ch. 2, I presume I can assume ch. 1 tubes as known good, right?

[pdf64]

...
Schematic is below.
 ...

Oh crikey!
That design is asking for trouble. Given typical valve guitar amp build practices of the early 70s (especially given the chassis wiring photo!), I'd be surprised if channel 2 wasn't a fair bit noisier / hummier than channel 1.

What surprises me is the frequency - are you absolutely sure the hum is 50Hz, rather than 100?

If the ECC83 in reverb circuit is removed, does the hum stop?

Replace the ECC83; how about if the ECL82 is removed?

How about if both reverb circuit valves are removed?

[pdf64]

Also which controls affect the hum level?
I suppose the channel VTB will?
But how about the reverb circuit controls -
intensity volume
intensity reverberation
volume reverberation?

Here's a schematic link, flipping between viewing downloads and a thread on safari is a problem.
https://el34world.com/charts/Schematics/files/Fbt/Fbt_500_r2.pdf

[Willabe]

Oh crikey!
That design is asking for trouble. Given typical valve guitar amp build practices of the early 70s (especially given the chassis wiring photo!), I'd be surprised if channel 2 wasn't a fair bit noisier / hummier than channel 1.

Ok, please explain. What do you see as the problems?

[pdf64]

Oh crikey!
That design is asking for trouble. Given typical valve guitar amp build practices of the early 70s (especially given the chassis wiring photo!), I'd be surprised if channel 2 wasn't a fair bit noisier / hummier than channel 1.

Ok, please explain. What do you see as the problems?

Channel 2 has all that reverb circuit inserted between its input sockets and input stage. So even with the 'volume reverberation' turned down , there's an additional gain stage in the channel 2 signal path.
Even with 'intensity volume' control adjusted for unity gain, that will tend to open the door to ground loops / reduce the signal to noise ratio.

[Willabe]

Can you please explain 'No'? I'm not doubting you, just trying to understand your reasoning.

I'm saying it might not be the heater string that's causing the humm. I doubt that because as I already said, if 1 channel humm's from the heater string, then both should humm. You think because the verb channel has an extra gain stage, that causes the humm in the verb channel, but not the other. You should still hear some humm in the other channel if it was the heater string.

And;

1. Like I said, it could be the way they routed the heater wires, lead dress, injecting noise into the tubes grid.

2. pdf64 could be right about the way they added the verb circuit.

3. It could be a bad/noisy tube. But if you have several of these amps and there all have pretty much the same humm in channel 2, then it's less of a chance it's a bad tube.

Also I just realized the ch. 2 gain and reverb tubes are on the side of the chassis closest to the mains transformer (preamp is in a separate metal housing from PI and power tubes). So I'll also try taking the amp out of the chassis and seeing what happens if I put some distance between them.

4. Yes, it could be the verb circuit is too close to the PT.

5. Or, it could be that the verb circuit itself because of grounding is hummy.

6. It could be the verb tanks output jack end, which has the lowest/smallest signal in the amp, is too close to the PT.

Do like HBP said turn the verb off for now, still humm?

..... I will try swapping ch. 1 and ch. 2, I presume I can assume ch. 1 tubes as known good, right?

Maybe, try it and see what happens.

[Willabe]

I'd try this; pull both verb tubes, ECC83 and ECL82, then use a gator clip lead to go from the verb input to the verb output on the schematic = where the 2 x 68K come together and pin 2 of ECC83a. That will take the verb circuit completely out of the amp.

If that stops the humm, it's something in the verb circuit.

[tubeswell]

If it’s in channel 2 only, with or without the reverb level up, then it’s probably around the first tube in channel 2.

[AlNewman]

So you have a bunch of these amps, and they all have the same hum, but you've done the same modifications to all of them?

In your schematic, there's a green arrow denoting B node as a power supply for the recovery and PI tube.
Did you modify the amp to run the recovery and PI off the screen node?

[dankinzelman]

So you have a bunch of these amps, and they all have the same hum, but you've done the same modifications to all of them?

In your schematic, there's a green arrow denoting B node as a power supply for the recovery and PI tube.
Did you modify the amp to run the recovery and PI off the screen node?

When I received the amps, they did not match the schematic. The green shows the differences between the amps and the schematic (and the work was very clean, so I think they came from the factory this way). Two of the amps have 5R resistors between the B+ diodes and the first reservoir cap. All three amps arrived with the PI powered off the B node, and there was a low level hum coming from the PI due to the dirty power. I moved the PI to the C node (as the original schematic suggests) and they became much quieter.

[AlNewman]

So if you unplug the trem tube, do you still have hum?  I wonder if maybe you created a grounding issue when you switched the PI and recovery from node B to node C.

[dankinzelman]

I'd try this; pull both verb tubes, ECC83 and ECL82, then use a gator clip lead to go from the verb input to the verb output on the schematic = where the 2 x 68K come together and pin 2 of ECC83a. That will take the verb circuit completely out of the amp.

If that stops the humm, it's something in the verb circuit.

Thanks. I did this, with the reverb circuit bypassed, it's dead quiet (but there's very little gain without the extra stage). The hum seems to come from the ECC83 in the reverb circuit. Turning up the reverb itself brings out only HF trash (maybe diode switching noise).

So if you unplug the trem tube, do you still have hum?  I wonder if maybe you created a grounding issue when you switched the PI and recovery from node B to node C.

Unplugging trem tube seems to make no difference. Moving the PI node drastically reduced hum overall (which was present even with all preamp volumes turned down), so even if that makes the preamp hum a little more, I think it's a worthwhile compromise.

[Willabe]

I'd try this; pull both verb tubes, ECC83 and ECL82, then use a gator clip lead to go from the verb input to the verb output on the schematic = where the 2 x 68K come together and pin 2 of ECC83a. That will take the verb circuit completely out of the amp.

If that stops the humm, it's something in the verb circuit.

Thanks. I did this, with the reverb circuit bypassed, it's dead quiet (but there's very little gain without the extra stage).

The verb channel, with the verb tubes out, should now have the same gain as the other channel.

Is it the same gain as the other channel now?

The hum seems to come from the ECC83 in the reverb circuit.

Why do you think that?

Turning up the reverb itself brings out only HF trash (maybe diode switching noise).

With the reverb tubes out?

[Willabe]

I'd try this; pull both verb tubes, ECC83 and ECL82, then use a gator clip lead to go from the verb input to the verb output on the schematic = where the 2 x 68K come together and pin 2 of ECC83a. That will take the verb circuit completely out of the amp.

If that stops the humm, it's something in the verb circuit.

Thanks. I did this, with the reverb circuit bypassed, it's dead quiet.

Now you have to find out what in the verb circuit is causing the hummm, could be a few things.   

Check for each of these. #4 is very important.

1. It could be a bad/noisy tube. But if you have several of these amps and there all have pretty much the same humm in channel 2, then it's less of a chance it's a bad tube.

Also I just realized the ch. 2 gain and reverb tubes are on the side of the chassis closest to the mains transformer (preamp is in a separate metal housing from PI and power tubes). So I'll also try taking the amp out of the chassis and seeing what happens if I put some distance between them.

2. Yes, it could be the verb circuit is too close to the PT. Did you try this?

3. Or, like pdf64 said, it could be that the verb circuit itself because of grounding is hummy.

4. It could be the verb tanks output jack end, which has the lowest/smallest signal in the amp, is too close to the PT.

[pdf64]

As asked a few days ago, 'are you absolutely sure the hum is 50Hz, rather than 100?'

[dankinzelman]

Thanks for all your help.

Here's some answers.

As asked a few days ago, 'are you absolutely sure the hum is 50Hz, rather than 100?'

I have more detailed info now after using a spectrum analyzer. Turning up volume on Ch 2 with reverb off gives me 50hz hum. Turning up reverb very high only increases the 50hz slightly, but brings out nasty harmonics at 150 and 250hz. These are actually much more annoying than the reverb, and I'm curious to know why I'm getting 3rd and 5th harmonics only from the reverb. Any ideas? Is it worth looking for a grounded shield for the ECL82 reverb driver?

Now you have to find out what in the verb circuit is causing the hummm, could be a few things.   

Check for each of these. #4 is very important.

1. It could be a bad/noisy tube. But if you have several of these amps and there all have pretty much the same humm in channel 2, then it's less of a chance it's a bad tube.

2. Yes, it could be the verb circuit is too close to the PT. Did you try this?

3. Or, like pdf64 said, it could be that the verb circuit itself because of grounding is hummy.

4. It could be the verb tanks output jack end, which has the lowest/smallest signal in the amp, is too close to the PT.

1. I will try some more swapping.

2. It didn't seem to make much difference.

3. How do I check this?

4. The tank output jack is on the opposite end of the chassis from the PT. I will try moving it further away with the amp disassembled, but the chassis design and umbilical only allow me a little bit of freedom of movement here.