New tubes and bias but loud hum

[birthcontrolrocknroll]

Hi, I have an Ampeg v50 with 2 12ax7's and 2 el34's that I just had biased and cleaned up when I got the new tubes. The only thing is is that now the hum is louder than ever before. Even at very low clean volumes the hum is loud. Can anyone tell me where to start looking for the problem? I still have a lot to learn about tube amps. Thanks.

[fixitrightthefirsttime]

Hum, or no Hum

some of the usual suspects for bad hum are:

1. positive bias on the 6.3 v ac filament voltage, such as from an open center tap on the filament section of the PT. the filament ac voltage is better off when negative WRT the cathode
2. weak electrolytic filer caps in the B+ power supply
3. short between the filament pins and the plate or grid pins on any socket
4. "high open" problems on jacks, sockets, solder joints
5. bad tubes
6. poor grounding

does the hum persist with all the knobs turned down?
does the hum occur with standby OFF?

[Rob Mercure]

Is the hum 60 cycles - like touching the end of a plugged in guitar cord with the amp turned on? Or is it 120 Hz - twice the frequency/octave higher? The latter points to something in the power supply and the former to heaters, bad jacks, or external noise getting into the amp.

Rob

[Mars Amp Repair]

new tubes & hum

as mentioned, I have also experieced this effect when replacing the 12AX7's in some of the old Ampegs. I suspect it had to do with the hum balance on the filiament windings.
I didn't have time to troubleshoot this amp, but just had to fish around & find an old 12AX7 that didn't produce a hum (ostensibly from the filiament to cathode voltage issue).
I wonder if anyone has experimented in a situation like this to try & raise the dc biasing of the AC filiament voltage of these preamp tubes in the old Ampegs. As I mentioned, this amp needed to be fixed quickly with only tubes, so I didn't have the opportunity to mess with it further. glen

[fixitrightthefirsttime]

Biasing the AC fiilament voltage

Putting a negative DC bias on the filament voltage (regardless of AC or DC heater supply) will, of course, be negative wrt the cathode. This is good.

On the other hand, a positive DC bias on the filament power supply is generally bad, especially for 6.3 VAC (unless you like hum.)

[Amp Kat]

I haven't experienced those preamp problems Mars with the older Ampegs but that doesn't mean there isn't something there. As long as the wires are twisted down the line using 6.3VAC there shouldn't be much hum. This method has been used for years with no problems at all if wired right and of course you can create a virtual center tap using 100 ohmers from each tap to ground . If the amp was not humming bad then, you had it biased and now it's humming then I would question the bias ? Are the tubes redplating ? exactly how much current is each tube drawing ? Are they matched tubes ? How old ? Is it possible that a ground wire came off when cleaning it ?

[R.G.]

Putting a negative DC bias on the filament voltage (regardless of AC or DC heater supply) will, of course, be negative wrt the cathode. This is good.

On the other hand, a positive DC bias on the filament power supply is generally bad, especially for 6.3 VAC (unless you like hum.)

Actually, I think that's reversed. You really don't want electrons emitted from the hot filament. If this happens, the emission is modulated by the filament voltage and causes... yep, hum. If you elevate the heaters positive with respect to the cathode, any electrons boiled off the filament are sucked right back onto it.

I believe there are a number of commercial hifi amps and some guitar amps with the filaments elevated.

[R.G.]

Hi, I have an Ampeg v50 with 2 12ax7's and 2 el34's that I just had biased and cleaned up when I got the new tubes. The only thing is is that now the hum is louder than ever before. Even at very low clean volumes the hum is loud. Can anyone tell me where to start looking for the problem? I still have a lot to learn about tube amps. Thanks.

A really good place to go look first is the Tube Amp Debugging Page at GEO -

http://www.geofex.com.

There is a complete run down there on how to isolate amp problems, including hum, which contains a lot of the tricks for isolating it.

[fixitrightthefirsttime]

(Orignally posted to the wrong part of the thread...)
Good point. It is true that some circuit designs need or even require an elevated "LT." In the final analysis, it probably depends on the circuit.

My expericence is that a positive (relative to the cathode) bias on the filament AC will, in some circuits, modulate the otherwise steady state as a function of the AC voltage. This causes a 60 Hz ripple on whatever is flowing to the plate, and voila,' the dreaded hum. This is especially critical in the preamp tubes, and I suspect in single-ended Class-A amps.

I saw a good example of this recently on a '67 Super Reverb with an open heater's center tap in the tranny. So, it floated to around +40 V bias and the hum was very bad. What was cool, is that we could balance the Normal and Vibrato volume controls to cancel the hum, seemingly from the tubes' heaters' relative phase. The output stage was matched and biased well, and that may be why we heard no hum at all with both volume pots down to zero. We had a chance to experiment with an external heater supply (before replacing the tranny) just to satisfy our curiosity.

I suppose that anything positive wrt the cathode will tend to draw current from the cathode, and this can upset the apple cart. If the filament is negative wrt the cathode, it will not encourage emission from the cathode.

I surely don't know all the answers, and am usually mistaken on some point or another on a regular basis every few milliseconds, so if someone can offer a different or better theory, I'm all ears.

[Enzo]

I would quibble with your analysis. Your '67 had an open CT on the heaters and floated up to +40 and hummed. So you blame the DC??? I'd blame the fact that it was no longer grounded, not the DC voltage. Hum abatement is the purpose of the center tap grounding after all.

And the fact that you can balance the two channels to cancel hum? I think the hum was induced into the preamp stages, and would have been regardless of the heater phase. In other words if you reversed the heater wires on one channel, it would still hum. Note the vibrato channel has one more triode in it than the normal channel does. SO the hum signals from each input stage are going to be out of phase with each other. Set them equal and they cancel.

Try some 12AX7s with the spiral heaters and see if they hum less.

I would not be too concerned with the tiny current emitted from the INNER surface of the cathode to the heater. I don't think it will affect the current from the working surface to the plate to any significant degree.

An old trick is to connect the heater CT to the cathodes of a cathode biased power tube pair.

[fixitrightthefirsttime]

I'm not suggesting that significant current necessarily flows between the heater and the cathode. It is the E field I'm talking about here. Is it not conceivable that the + biased oscillating electric field will modulate the steady state of the triode? In contrast, a neg biased oscillating field would have less effect. The system is already biased to the acceptable idle state. My hunch is that during a good signal, the neg biased ac WOULD modulate the system, but the signal-to-noise ratio is acceptable. Clearly, hum is more noticeable when the guitarist is tacet.

Rhetorically I ask, what is the difference between an AC signal with no CT which floats to a, let's call it "virtual bias," a stable point dictated by the physics of the situation, and an intentional DC bias? Either way, it is oscillating around that DC voltage. It is measureable, repeatable and reproducable. And curing that makes the hum problem vanish.

Yes, yes. Grounding the CT eliminates the problem. And you are suggesting there is a 180 degree phase differential in the heater current between the two channels, and the gains of the two channels can be used to cancel it...mathematically the same result as in a balanced push-pull cancellation? Plenty of whisper quiet PP-A/B amps hum like crazy if you pull one side of the output tubes. I think we're trying to make the same points, or?

[Enzo]

The PV 5150 for example sets the preamp tube heater 6VAC at +24VDC. That amp is known for its hiss due to the very high gain, but it is not known for hummy.

"Rhetorically I ask, what is the difference between an AC signal with no CT which floats to a, let's call it "virtual bias," a stable point dictated by the physics of the situation, and an intentional DC bias?"

The difference is one is floating and the other not. Your floating 6v is not referenced to ground. You can reference it to ground directly with a CT, or directly with the pair of resistors virtual center tap, or you can reference it indirectly through the low impedance of the DC supply. The 6VAC signal is there no matter what, it is the grounding of it that reduces the hum.

There are two issue here: referncing the heaters to ground one way or another, and whether elevating them DC makes a difference as well. The heaters need to be grounded for lower hum. Then elevating them might tend to reduce it further.

In fact even simply grounding one side of the 6V is better than floating.

"And you are suggesting there is a 180 degree phase differential in the heater current between the two channels, and the gains of the two channels can be used to cancel it...mathematically the same result as in a balanced push-pull cancellation?"

NO, you are putting words in my mouth. The two channels are 180 degrees out of phase. The input stages are the same, so the hum in them is too. It cancels when you mix one with the other.

I am not sure what the PP stage example means. Of course if you yank one side of a quiet PP stage the hum will increase. But if you adjust the two preamp channels in your example so they cancel and then pull one of the input tubes, voila it will hum too.

[Amp Kat]

The PV 5150 for example sets the preamp tube heater 6VAC at +24VDC. That amp is known for its hiss due to the very high gain, but it is not known for hummy..

That's actually 48 VDC +/- 24 but same principle and that elevation is for the first preamp tube and the splitter. No elevation for the power tubes and no need for it. The humbucker effect works great in power tube hum cancelling and many Silver faces reverted back to blackface standards to eliminate it. Bias-balance controls will eventually drift away from it's center causing hum. Now preamps are a different story and the tubes that go with it mainly 12AX7's with the higher mu amplify the hum more. You either have a low hum tube or not and there are extreme sonic differences in their hum levels from tube to tube and type to type. I have noticed big differences in elevating Mic Pre amp filaments over 50 volts Dc but a series connected 12VDC may be just as well mainly on that first tube and many amps like the JCM 800 and 900 series DC the preamp tubes with a rectifier coming from the power tube array with the exception that they use a poor BRD rectifier and it burns up.
But many amps use elevation with low hum and achieve lower buzzing from preamp tubes

[Enzo]

KB, thanks, you are right. I mis-spoke. The 5150 lifts the POWER TUBE heater 6VAC to +24V. The preamp tubes are indeed on DC in series string. I grabbed the first example to my mind, but it was not the best choice.

For reference see the 5150 schemo power supply page wherer the FIL supply comes in. R69 pulls it up.

[fixitrightthefirsttime]

I need to catch up on repairs, so I'll have to go on the wagon insofar as forums and freecell for a couple of days.

It would be nice to get back to the original argument, however: are we of different opinions whether a negative dc bias to the filament voltage will encourage hum? That was the first discusssion.

I only know of two recognized ways to resolve such an issue: theoretical or empirical. And theory should be "not opinions," and use only the most basic laws, at least in the beginning. As long as we limit our discussion to what happens after the electrons leave the surface of the heated cathode, by definition from the solid state to the plasma* state, the four relevant basics here are Maxwell's equations, which you probably know covers everything, literally, in the classical physics of electricity and magnetism. I suggest we start with Gauss' law. Obviously I mean the "del-operator scalar product with the E vector," NOT the "conservative-field closed-path integral E vector dot-product 'dl.'" And certainly everyone knows the former is "rho" and the latter is "zero," right?



*in the real meaning of plasma, not the "common" connotation more accurately described as "glow discharge," such as that observed in aurora, nebulae and lightning, or that which is used in pool hall neon bud lite signs and overpriced tv sets on which we can watch leave it to beaver reruns.