Bassman AB165 low level noise issue

[mresistor]

There is some static sort of hiss with kind of a crackling type sound coming out of this amp with no input at all, and the volume turned up a little. Both channels. You have to put your ear up close the speaker to hear it.  I've read this is a common issue with these older amps. I changed out every tube one at a time and to a brand new duet of 6L6s with no change.  Am I correct to think that this could be one of the carbon comp. 100K plate resistors on the small bottles? I pulled the tubes for each channel separately and it makes no change.
There are three carbon comps that had been changed in this amp. Two were the 1k5s 6L6 grid stoppers and one 1k5 in the cathode bypass on the first stage of the bass channel. The others are all original, except maybe the one on the bias board has also been changed at one time. 
I have replaced three 20uf filter caps in the PS. I have changed the bias circuit to regular bf style. I replaced the .022 coupling caps with new .047uf caps. I also changed the bass and treble caps in the bass channel tone stack to .022 and the slope resistor to 56k. I disconnected the death cap (amp has three prong power cord) and reconnected it and it makes no difference in the audible noise.
I checked the grounding of the input jacks and pulled them to check for corrosion, they look good. They check good. I haven't cleaned the contacts yet. Would that cause this noise? I fixed the newer ground wire connection from the power cord. It was just bare end of wire under one PT bolt, I soldered on a connector and then bolted it down.
Also   I suppose this could be corroded tube socket pins and what would you clean them with?
Any other insight?
BTW since changing those 3 filter caps, changing the bias circuit and fixing the power cord ground, the plate voltage has dropped 10v to 442v.
Oh  and I have removed the local feedback resistors from the 6L6s.

[mresistor]

And  when the volumes are turned up the noise is also louder.

[HotBluePlates]

... corroded tube socket pins ... what would you clean them with? ...

I would clean tube pins (if needed) and socket contacts with DeOxit.

... when the volumes are turned up the noise is also louder.

You could assume the noise is before the volume controls then, which puts it at the 1st gain stage for each channel back to the input jacks. You can try replacing carbon comps to get rid of it (because given enough time, all carbon comps will become noisy), or you could use a Listening Amp (bottom of the page) to localize the noise.

Plate resistors are usually blamed (because they often become noisy), but at least one forum member fixed a noise like this which turned out to be a cathode resistor. The Listening Amp would help you find this, though you have to be cognizant of how noise in a grid or cathode circuit gets amplified by the tube and sounds louder at the plate.

[mresistor]

I re-read the instructions on the listening amp. It looks to be a useful tool. 

[mresistor]

Also  the deep switch is popping really loud when switched the first time.

[HotBluePlates]

Also  the deep switch is popping really loud when switched the first time.

That is normal; the pop happens when the cap charge from 0v to some other d.c. level.

Try installing a 4.7-10MΩ resistor across the switch lugs. That should neutralize the popping.

[mresistor]

Cleaned the sockets and tube pins.   still there

[HotBluePlates]

Which is why I posted about the listening amp.

Many things can cause noise in an old amp. You can waste your time & money guessing/replacing, or use a tool to find what's making the noise. Your call.

In the end, the "listening amp" is just a probe, cord and a cap. You supply an existing audio amp which hopefully doesn't contribute much noise of its own.

[mresistor]

What about a handheld O'scope?

[HotBluePlates]

Give it a shot. You'll have to pay careful attention to estimating an average noise level for each area you probe. Also be thinking about the fact that a noise at a tube plate may look bigger than a noise at a cathode or grid, but that may just be due to the gain of the tube.

[mresistor]

Ok I built a listening amp tool into a single gang jiffy box.  Now working on making a shielded wire with a probe to fit a phone jack.  I have a quiet champ amp so that is what I'll be using as the listen amp.
More fun than anyone ever expected...   I would have to buy a handheld Oscope and I'm seriously thinking about it.  I thought of trying an app for my phone but I don't want to risk destroying my phone. 

[mresistor]

Got the listening amp together.  Hooked up, and probed around inside the amp. Discovered that crackling noise is present at pin 2 of the PI (12AT7).  Its also more pronounced, louder, at the 47K feedback resistor. So it looks like it is coming from the OT where the feedback resistor takes off.

[mresistor]

Reflowed the solder joints for the feedback wire at the speaker jacks, and then on both sides of the 47K resistor. Hooked the amp up to my 2x12 cab and gave a listen. The crackling noise is pretty much gone now.


Thanks HPB for the kicking me in the butt to build and try the listening amp tool. It's a pretty good tool.

[6G6]

As stated before if it is louder when you turn up the volume
it is before that.
Replacing the old carbon comps in the first stage would cost very little time and almost no money.
Maybe that will do it, maybe that will rule out one likely source and give you a little piece of mind
knowing it's not that.

[HotBluePlates]

Awesome!

6G6 is right though, changing noise with changing the volume control position indicates a noise before the control. So perhaps something intermittent (which is typical of noise in old amps).

It's your call whether to wait or to proactively replace parts.

[mresistor]

I think it sounds pretty good ,now...  even turned up now that crackling noise is way less.  If it were my amp I would replace all the 1.5K cathode resistors and all the 100K plate resistors, since I have some precision 5 bands 1% metal film resistors sitting here. But I don't want to get too carried away with replacing stuff and let the owner take it and play it as is  and get some feedback from him. I think the bass channel is totally usable now, where before it was pretty dark for guitar. I also read that these early BF Bassman amps are notorious for being a little noisey. So, I'm going to wait and see what he has to say.


Thanks again guys.

[mresistor]

Ok the owner and I evaluated it and decided to try to eliminate most of this noise - IOW try to get the amp as quiet as we can. I am going to try 6G6's suggestion now.
6G6  are you talking about the 68K and 1M on the input jacks? Also there is a 100K plate resistor, and the cathode resistors on the imputs.

[HotBluePlates]

Ok the owner and I evaluated it and decided to try to eliminate most of this noise - IOW try to get the amp as quiet as we can. I am going to try 6G6's suggestion now.
6G6  are you talking about the 68K and 1M on the input jacks? Also there is a 100K plate resistor, and the cathode resistors on the imputs.

If the noise level changes when the volume control is manipulated, then anything before the volume control is a suspect. That might be the tube itself, resistors on the tube stage or input jacks, 100kΩ resistor in the tone stack, etc. It could even be a dirty tube socket or dirty contacts on the jacks.

If you want to make minimal changes and/or know exactly what caused the noise, you could replace resistors one-by-one with metal film while listening after each replacement. If you're gonna be building or working on Fender-style amps in the future, it's probably not a bad idea to have a stock of resistors in each of those values.

Usually, you'd suspect the highest resistance first as being the biggest noise contributor (in fact, self-noise of a tube or stage is sometimes specified in terms of the "equivalent noise resistance"). But your new listening amp is probably very helpful here for places to attack.


Alternatively, you might say the carbon comps are suspect and will be a noise issue now or in the future and replace them all. That's your call. I'm doing a build now where for my own curiosity I'm trying a mix of wirewound (wherever possible) and metal film, as wirewound resistors are the quietest type available. But I have a prior build with almost all metal film that sure seems noise free to me.

[mresistor]

Thanks HPB - I have a bag of precision 1% 1w 100K metal film and one of 1.5K. I have some assorted others but not in half watt. I have some carbon films in 1w which are the size of 1/2w carbon comps. The 1w metal films are about that size too. Then I have larger ones  2w and up.   
Think I will start with the 100Ks on the plates  one at a time.  I sure hope it doesn't end up being a tube socket, but we've seen that here before. It's really the last thing one would expect, I don't know why; maybe because they are so long lasting and reliable usually.


here is a sample of the noise at the present time   volumes  up around 8-9    both channels
http://vid122.photobucket.com/albums/o250/privateacct/MVI_4122_zpsvbdhcspo.mp4

[mresistor]

I replaced all three of the brown turd caps in the feedback loop and front of the PI and the crackle type noise went away.  What I hear now is like white noise and a little 120hz hum I think and that hum was the normal channel.

[HotBluePlates]

Keep trying to replace resistors and see if you can knock the noise down.

Do you have any tubes you're using in a known-quiet amp? You may try swapping them in to the Bassman for a reality-check to rule out a hissy tube.

[mresistor]

Thanks HPB - I used very quiet tubes to check already from a very quiet amp. Did not make a difference .

[HotBluePlates]

There was no way to get a frame of reference for how much lower the noise was compared to the amp's sound at that volume level.

Listening to just the noise, which may have been normalized to a high level, it sounded like a hiss machine. But it might not sound that way in person. Or after you've played for a few minutes at 8-9 and you're ears are desensitized...

[mresistor]

I changed the AB165 feeback resistor which is 47k to the AA165 value of 820 ohm and that almost completely eliminated the noise issue.
It also seems to have reduced the apparent power out.


It sounds 100% better now without all the noise and hum.

[mresistor]

Still though, the amp does not have the output it should. I wired in a pot to see what value of fdbk resistor I could use while still keeping noise to a minimum, and I came up with 2.8k ohm. So I tacked in a 2.7k resistor. It has more output and is fairly quiet. But still does not have the output level a 50w blackface fender should have.

[mresistor]

Ok, now I am getting somewhere. I lifted one end of the 470K local feedback resistor on the 3rd amplification stage, where both channels go in this AB165. It really unloaded that stage and the whole amplifier output increased.  Still a little shrill at higher volumes,  so I think I have to fiddle with the value of the global feedback resistor again before the PI to sort of find the sweet spot. It now has BF tone in spades, and is still pretty quiet.