• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Hummm...

I may do that, at least all the ones I can get to reasonably. I did get a few minutes this evening & started to collect what I have. I also did the following"
1. grounded the chassis to the ground in the electrical socket. Made no difference.
2. Replaced the 2 6AN8's with 2 Sylvania tested & supposedly matched ones (originals were Heath branded Sylvania's too). This made the hum worse.
3. Replaced the 12AX7's with a pair of decent-testing Telefunken's I had. Made no difference.
4. Swapped the 2 6AU6's. Made no difference.
 
Well, I replaced all but the tone control resistors with 1/2 watt metal film ones. I did check the values before installing the new resistors. The resistors around the tone controls were values I did not have, so I un-soldered one end & tested them. None were grossly off, the worst one was 12% high (replaced), several were 8 - 10% high. If anything, it is slightly noisier now. There is more volume-dependent hiss. Low-level hum is about the same. Waves sound maybe a tad better but not much. Any other ideas? Could I have damaged a cap somehow? I haven't swapped out the the 6AU6's, 6BQ5's, or the rectifier as I don't have suitable replacements, but all seemed to work originally except the rectifier (it was dead & I had to buy another), and all tested good.

I am thinking I did something wrong when I re-capped it, and that I need a fresh set of eyes to double-check my work as I don't see anything. I really feel like I am moving backwards on this one.
 
The fact that the 120 Hz hum is gone says to me that the recapping is fine. 60 Hz hum has got to be coming from something that can induce 60 Hz hum. From my experience this can come from filament circuits, filament voltage bleed through to the cathode of one or more of the tubes, wiring layout (lead dress), ground layout, or a ground loop. I think it is likely that this 60 Hz hum has always been there since the amp was originally put together, so my take is you have not made things worse.

To hone in on this hum, as I eluded to previously, one thing you can try is to remove all the tubes except the output tubes, rectifier, and the 6AN8's. Thus remove all of the preamp/frontend tubes. Then turn on the amp and make an assessment if it still hums, adding back the previous stage's tubes stage by stage and retesting for hum after each stage's addition. Find what stage causes it to start humming. That will hone you in on the stage to focus on. This is kind of a poor man's test, but lacking any other ideas, it's not a bad place to start. (I make that recommendation after studying the schematic to know that each stage is cap coupled, so you can remove tubes from stages without disrupting the operation of the stages following. The only exception is the 6AN8 is direct coupled between the pentode voltage amp and the triode inverter, but since those stages are inside the same glass envelope, it's okay to remove even the 6AN8 if you want).

These are just ideas. I've been there before and sometimes it's hard to see the problem because you've stared at it too long.
 
Well, it hums with no tubes in it, just not quite as loud. Removing any pair of the 6AU6's, 6AN8's, and 12AX7's did not make it go away. Removing all 4 output tubes did not make it go away. But with all the original tubes back in their original places, the hum was stronger in channel A. In fact, the residual hum in channel B was barely noticeable, it might have been there all along. Swapping the output tube pairs moved the bulk of the hum to channel B. Replacing the "hummier" pair with a completely different pair of tubes made more total hum with the bulk of it in channel B (new tubes). So then I tried swapping the one tube from each original pair to the other channel. I could not isolate a "bad" tube, and the hum never went away or changed level much, it just changed from one channel to the other with certain pairings. That makes me think that part of the problem is output tube (or output tube matching), some is not. The ocean sound is pretty much gone, but there is some cracking sound as the tubes warm up.
 
If you have hum with no tubes in there it might be coupling between the power and output transformers. Might be worth looking at the orientation of the transformers. The power transformer and the output transformers need to have their cores oriented so they are at right angles. If someone assembled the kit wrong and got the transformers clocked wrong it might be the cause of this.


An extremely dirty power supply could also do it. Check the first filter cap's ground. It should be grounded to the same point that the center tap of the power transformer is.
 
The transformers cannot physically go in 90 (or 270) degrees off. They could be 180 degrees off but that would not change the alignment of the cores, right? The first filter cap grounds through the can body, and there is a center tap wire from the power transformer to a lug on that can. Would it make sense to try putting another cap there as a test? If so, what size?
 
Last edited:
Just to see, I disconnected the first stage of the can cap & replaced it with a free-standing 68 mfd 400V individual capacitor. I still get the same low-level hum with the output tubes out.
 
Ahhhh. Somehow I missed that it is humming w/o power tubes. Agreed, then it must at least partly be due to magnetic flux spilling over from power to output transformers. It could also be very close proximity of the power cord to one of the output transformers. Just out of curiosity, is this a steel chassis? Steel can harbor eddy currents which could contribute. Some power transformers utilize a Faraday shield around the windings. (some Fisher integrateds for example) This is something like an inch or two wide copper band around the center of the transformer, under the end bells. The copper band is grounded via physical contact with the end bells to the chassis via mounts.
 
Ziradog said:
The transformers cannot physically go in 90 (or 270) degrees off. They could be 180 degrees off but that would not change the alignment of the cores, right?
Click to expand...

Right. ok, well so much for that. I was thinking maybe by some odd chance it had transformers that could mount in any direction. Looking at pics, I see they are rectangular. Incidentally, it sure looks like the cores lay in the same direction, which does not help. There is a Faraday shield on the power transformer though, or at least there is in this picture I found on google. If that has it's own lead out, make sure its grounded. Sometimes its just bonded to the frame and grounds through the mounting bolts, sometimes there is an "extra" wire, usually of fairly light gauge that connects to the shield. If its just by physical mounting, maybe pulling the transformer loose and wire brushing those mounting points and hardware wouldn't be a terrible idea.

The first filter cap grounds through the can body, and there is a center tap wire from the power transformer to a lug on that can. Would it make sense to try putting another cap there as a test? If so, what size?
Click to expand...

That sounds reasonable. Since you've already tried another cap, I guess that wasn't it after all. Just trying to think of things to try here. Kits can be a pain since you don't always know if problems are shortcomings in design or original build. Some Google foo tells me maybe its at least partially the design since a search for "Heathkit SA-2 hum" gets a lot of results from people who seem to have done the normal stuff but haven't been able to make it shut up. Some posts do imply that the wire lead dress is very particular, so if you can dig up an original assembly manual to confirm that things are routed like the instructions say, that would probably be helpful.
 
I will check that strap tomorrow. I don't recall a lead. I have a print out of a manual but the pictures are pretty crappy. The wiring looks like it follows the correct routing to my untrained eye. And realistically, the hum I get with the tubes out is livable if it doesn't indicate other issues. I do plan to ad the extra filter stage that Dave Gillespie suggested elsewhere. Any thoughts on the noise with the tubes in? It was suggested that I buy another set of cheap tubes, but even cheap tubes aren't real cheap, I would rather spend money on good matched tubes but only if I have good reason to think they will help.
 
I played with the Faraday strap, made no difference. I even connected it to the ground rail with a jumper, no change. Someone asked is the chassis was steel, it is, but there is a rail that everything is grounded to rather than the chassis. The wiring appears to all be routed correctly, twisted where the manual says it should be. The loudest portion of the hum is concentrated in one channel when the tubes are in. It gets louder as the tubes warm up, then tapers to a lower (but still annoyingly loud) level over ~15 seconds. When the tubes are out, the hum is less. There does feel like there is some vibration when I touch the transformers/can cap/chassis, but I can't isolate it (seems the same at any spot). Not sure what that is telling me, but the hum I hear is definitely coming from the speakers, not the unit itself. I ran it at a fairly loud volume (knob ~1/2 way up, loud enough in the garage for me to listen to it in the driveway while I did car work) for 1-1/2 hours, nothing felt hot (except the output tubes) or made any bad noises. Hum level was the same at the end as when it started. Still thinking I may try to find someone knowledgeable to recheck my work & see if I missed something. Any other ideas?
 
Really interesting article. As best I understand it, the ground set up is completely different than on the SA-2, and the B+ filtering is better as well (but not as good as the finished unit with a choke). I don't want to remove some of the stuff Dave did in that tone as I am not using a separate preamp. Moving some of the 120-volt stuff may be possible. I already ditched the phase reversal switch as mine did not work. I am also thinking I need better balanced output tubes for the reasons he has noted. A lot of the rest was over my head honestly.
 
Last edited:
So let me summarize what I think are the critical points of understanding related to nixing hum from Dave's thread, which you might then consider as you tackle the hum on your amp. You may already have some of these applied on your amp.

1. There should be one connection point only from circuit ground to the chassis.
2. For any given tube stage, connect the grid ground reference to the same point as the cathode ground reference. This eliminates potential for mini ground loops between the grid and cathode for that given tube.
3. Use RCA jacks that are insulated from the chassis.
4. Ground reference the filaments via a voltage divider from a suitable source such as the cathode voltage of the output tubes. This back biases the diode condition that exists between the heater and cathode elements, and thereby blocks any hum that otherwise might be generated by leakage through that path.
5. Magnetic energy from the power transformer can easily spill over into the audio circuit ground, and that coupled with a steel chassis can push current around the ground circuit, causing hum. Dave eliminated it by bringing a very low impedance wire (ie thick) from the left channel ground up out of the chassis and then back in to the right channel ground. I.e, he eliminated it in my understanding by removing the ability of the magnetic energy under the chassis from influencing the ground potential from left to right chnnel ground systems, and thus eliminated the ground loop potential there. Granted this step was quite specific to the AA-30. But the learning here I think is that ground loops can be induced magnetically and not just electrically.
 
Last edited:
That helps me understand. #1 &# 3 are done. #2 I think is done but will need to check. # 4 I can check reasonably easily, pretty sure it would take changing some things in the circuit that I don't really understand. #5 will take some thought.
 
I agree #5 is hard to get your head around.

But while all of that may eventually help, because of the fact that the hum in your amp occurs without tubes, it leads me to believe it's interaction between the power transformer and output transformers. Without being able to physically change orientation or position of the transformers, I don't know how else to eliminate it. Maybe through some sort of more aggressive shielding around the power transformer. Or maybe if you can address all of the other potential hum problems (if there are any), where the only hum left is the transformer interaction, it will be quiet enough to be tolerable.
 
Last edited:
Happy Easter!

OK, #2 is proving more difficult to check than I figured. EVERYTHING is grounded to a central rail, but not always to the same point on the rail, and some things are daisy-chained to the rail. I assume I want the grid ground & the cathode ground for a given tube stage to be grounded together to the rail, correct? So can anyone take a look at the schematic (link in message #26) and give me an example (like, on V1A, the side of R7 away from pin 2 and the side on R8 away from pin 3 should be grounded together)? I just want to make sure I understand what I am doing. I am ignoring #4 & #5 for the time being. Also, if I switch to a grounded power plug, where should I attach that ground? And any suggestions on how to bring the power cord into the chassis without cutting up the metal? The temporary ground does seem to cut down on some of the volume-setting dependent noisiness when the power plug is in one orientation.
 
Last edited:
Daisy chaining to the ground rail I think is okay, if that means a local star ground that then connects to the main ground rail. The important thing for this step is that the grid ground reference and the cathode ground reference connect at the same point.

Examples:
V1B: Bottom sides of R12, R13, and C7 connect to the same point.
V6A: Bottom sides of R21, C15, R24, C16 connect to the same point.

The ground of the grounded power plug should connect at the same point to the chassis that the circuit ground connects to the chassis. But again, the main problem at least from my distance appears to be the magnetic interference between transformers.
 
It looks like nearly all of these connections are set up as you described. There was one on one of the 12AX7's that was not, so I relocated that ground point. Doing so added some noise (buzz) so I will probably move it back. So items 1, 2, & 3 on your list from Dave G's post are all done. No idea how to do #5 yet. I also swapped out the rectifier tube with a loaner just because I had not had one to try before, that did not help either. Hopefully tonight I will add the additional filter to the power supply as suggested by Dave G. in some other posts, but I don't think doing so will help this problem. I am still trying to figure out how physically to hook up a grounded power cord without drilling a large hole in the chassis.
 
I don't know of any other way to do it; you need three wires coming into the chassis (hot, neutral, ground), not to mention a way to protect the insulation from being torn up by the sharp edges of the chassis, as well as strain relief. Although not what you want, an IEC plug would work, but requires a square hole.
 
You must log in or register to reply here.
Back
Top Bottom