I have been working on an RCA SA-3 amplifier console. It's a model I've always wanted since I saw one of Doug's posts about one a few years ago. I've had some frustrating hum problems with it.

The guy I bought it from demonstrated it to me to show me the transformer was good before I bought it, so I knew it sounded awful and needed a full recap, which I performed. I first did the recap of paper caps, then tried it, and it sounded great except it had a hum-- ostensibly, I thought, from the electrolytics. I rebuilt the cans as usual, cutting them, gutting them, and installing new caps soldered to the original connections. I turn it on and hummmmmmmmm.

I figured I made an error in the connections so checked, rechecked, triple checked.....no luck, still had the hum. Not a terrible hum, it was mostly masked by input volume---but at zero volume very noticeable and annoying. I have recapped similar twin-6V6 phono amps and never had this issue. I checked for lead dress issues but could not see any obvious problems there. I was working on it for days, checking and rechecking, testing all caps more than once....no change. I had not sealed the restuffed electrolytics so I was able to check all connections for shorts, etc.

Then I decided to take the restuffed caps apart and just mount them under the chassis as a last resort. The hum vanished. I then proceeded to push, prod, bump, shake the amp while running to see if there was any change, but it sounded perfect. It's back together now with those caps under the chassis and the empty cans just crazy-glued back on. I'm happy, because I love the amp, but I'm still puzzled. Could there have been some conductance via some traces of electrolyte that I may have left in the cans?

Anything in the cans causing the problem is extremely unlikely. I suspect you just ran into and accidentally cured an extrememly aggravating nemisis of technicians everywhere. A ground loop problem.

Back in the 70's I ran into this in some warranty repairs for some new stereo consoles. One channel had the low level hum that would not go away. The other channel was perfect. The solution after 2 days of beating my head against the wall was to disconnect the ground of one power supply capacitor, move it 1 inch to the left and reconnect it. Hum gone. There were no defective parts, no bad connections, no cracked circuit track, it was just a ground loop problem. Three more warranty repairs for the same console came in that summer and they all got the same 5 minute correction that worked on every one of them.

Curtis Eickerman

I am going to say you had a placement issue. Stuffing the electros in the cans, something was touching or amiss. If one is not extremely careful of the position drilling the holes you can run into the old leads in that waxy base or ?? I think your moving the caps just happened to resolve that issue. The cans being stuffed was not the issue it had to be something else. Remember the original electrolytic can is basically 3 or 4 caps built from one unit completely exposed to each other inside and there is not issue with that. So there is no way two sealed individual electros should cause issue properly placed next to each other.

I know the tube stuff I got from you all had the electros under the chassis so maybe this was a first time stuffing cans? One has to be very very precise when doing it or these issues can happen. Did you clean the old cans extremely well before stuffing them? What did you use to glue them in? Some adhesives may be very conductive as well. Also I always strip some 22 gauge 600V wire and use the insulation over the cap leads where they go through the can base. I do NOT use shrink tubing as it is far too thin for proper insulation.

BTW I absolutely love that SA-3...that is killer....I want one of those! Anybody got one out there for sale or swap? I got some cool stuff to trade! Show us some pictures of the back of that unit with the cover off.

Thanks for the answers, guys.

No, BB, I don't always stuff cans---depends on the project-- but I've done it a lot of times in the past 20 or so years of being a hobbyist, so it's not the first time. Believe me, I checked everything on this one, and a bunch of times. The only thing that I did not do that you mentioned was use the larger gauge insulation on the wires going through the can....I just use shrink on them, however, they are going though a non-conducting material so I didn't think that'd be an issue. I hadn't glued them yet, as I was still testing, so conducting glue would not be the issue. Plus...the same hum existed before I restuffed the electrolytics anyway. I certainly would not rule out some kind of mistake on my part, just can’t figure out what it might have been. Good chance it was a placement problem, but I may never know.

The ground loop issue that Curtis mentioned sounds good, especially since the hum existed before I replaced the cans. Even if it isn’t that, it’s good to hear that from someone, since it will give me another weapon to use against mystery problems in the future. I wouldn’t mind hearing an explanation of why a ground loop occurs, if anyone can expound…

Another thing that I found fascinating about this unit is something that goes back to the previous post where I first saw one of these. I believe it was Doug who said that these were made from “leftover” record player amps from the 1955-1958 mono era New Orthophonic record players. Well, when I got this, it appeared untouched, however, there were two soldered points that looked like they’d had factory wires going to them that were just cut and removed. There was still a nub of wire left that even still had its insulation on it. I see that as kind of proof of the “recycled amp” theory.

Well, if the hum was there before you re stuffed the cans then that would not have been the issue.

Here is a very crude explanation of a ground loop in very basic terms. The different currents in an amp always travel in a circle or loop from positive to negative. Depending on the placement of these currents circuits the can cross over each others path traveling to their negative point. When these currents cross each others paths you can get unwanted noise from each other. I am sure someone can elaborate much better then that but there is a simplified example.

If you want info galore about Ground Loops go to some of the Guitar Amp building forums and do a search. You will find TONS of great info. Ground Loops are HUGE problems with guitar amps because they run so hot with such high gain. They pick up noise from EVERYWHERE and ground loops are a huge problem. Just using the chassis as a ground like done in most hi-fis is a huge no-no as it creates all kinds of crossing ground loops. In Guitar amps the designers go to great lengths to keep the different ground loops from crossing each other and causing noise and hum issues.

From what I have "heard" most hi-fi and record player type stuff is not greatly affected by ground loop issues like guitar amps are because the Hi-Fi stuff usually runs with much lower gain. If there is an issue with a ground loop on a hi-fi it would either have to be a design fault in the circuit or someone has been in there and moved stuff around. RCA would not design a ground loop issue into one of their units, and if it was a circuit problem then all of these amps would have that hum issue. It would have to be an someone poorly placed a lead during a repair or it was assembled incorrectly from the factory.

Brain, isn't that an RS-164 amp (the same amp as SHF-6, SHF-7 and HFP models use)? I had a hum in one of these amps about a month ago after installing new electrolytics underneath on a terminal strip. I had the electrolytics grounded to the chassis, but not grounded "good enough." I put lots more hot liquid solder on the ground and really put it on thick, and the hum issue was cured. That may have been your problem as well.

On the contrary, ground loops are extremely important in hi-fi equipment. Real hi-fi components, particularly those with magnetic phono preamps, have gain values much higher than in guitar amplifiers. Grounding is a critical design element, and the main idea is to keep low-level signal return paths separated from high-level returns as well as from the filter capacitor ground return. Often you will see a heavy ground buss to which the signal returns are tied, which in turn is tied to ground at one point and one point only, to eliminate ground loops.

Even a lowly amplifier like that in a 45-EY-2 has a good amound of gain, and grounding was a design issue even for it.

It is very important when replacing electrolytic capacitors to make sure that the negative leads are connected exactly where the originals were. Often when replacing filter cans with under-chassis capacitors there will be the temptation to connect the negative leads of the new ones to any grounding point that's convenient and looks nice, but don't do that. Connect them to one of the lugs on the can. That's one reason I'm increasingly re-stuffing cans -- I am sure that the grounding will be right.

-David

David, on a transformer operated amplifier like these RCAs have, it is true that the electrolytics are tied to chassis ground through the grounding lug on the can. If one is re-stuffing a can, then the ground should be tied to the original grounding lug. However, if new electrolytics are installed underneath, either direct or on a terminal strip, they can be grounded to ANY good chassis ground point. I've been doing exactly this for over 3 years and these amplifiers are quiet as a mouse with absolutely no hum.

Are you sure about that David? I know you have quite a bit of expertise with hi-fi stuff but I think you are off on the comparison between gain in hi-fi and guitar amps here. Have you had a lot of experience with guitar amps? Maybe my wording was wrong someplace?

All I know is in tube guitar amps there are multiple stages of gain that can be called on if one wishes to put the amp into distortion by over driving the pre-amp or the output tubes. These amps usually have mutiple pre amp tubes to achieve this, far more then you ever see in any Hi-Fi amp, sometimes up to four (and I have heard of even eight) 12AX7s just for gain! I have never heard of a hi-fi amp that is designed to overdrive the signal into clipping with multiple pre-amp tubes or by running the out put tubes anywhere near the voltages of what guitar amps do. I know when I have built guitar amps off Hi-fi amps the one mod that always needs to be done is additional pre-amp tubes to increase the gain to the level needed to amplify the guitar signal. If you plug a guitar straight into a hi-fi amp you can barely hear the thing even at full volume....that sounds like far lower gain to me.

I have spoken with several very experienced techs and builders about the subject including DVC and folks who are actual guitarists as well as hi-fi experts and we were on the same page about the mater. In fact I just spoke with someone about using a guitar amp as a signal tracer because of the high gain abilities of it.

I know you have probably forgotten more then I know and you actually have me questioning myself over such a basic thing (LOL) but I have to disagree with you on this one or maybe I am misunderstanding something.

Maybe someone else with some expertise in the matter on both style amps can chime in out there....VanCleave you out there?

Well, I'm learning a lot here guys, so keep it coming!

....I just did a quick Google Search all I find is

Guitar amp Gain > Hi-Fi amp Gain

If you had a nice 60 cycle hum, then likely one or more of the e-caps were shorting out against the metal can you stuffed them into. I had that happen with a stuffed can once: the e-cap pulled away the insulation from the inside of the can, shorting the positive lead to ground. In that case, it was like taking it out of circuit, so it didn't hurt the cap, but introduced a wicked hum, as the power supply was no longer filtering.

I kind of stay away from restuffing cans unless there is no room under the chassis. I can only imagine that it would hurt the value of the radio, at least years from now when all of our repairs have failed and someone is looking over a set to purchase/rerestore, only to see that some key parts have been hacked up irreversibly. At least if it is under the chassis, if someone doesn't like it, they can cut it out.

This is exactly what I thought probably happened....

Larry, just because you have gotten away with haphazard grounding doesn't mean that it's a good thing to do. It's a poor practice which hasn't yet bit you in the a** because you are not dealing with a very high gain amplifier. RCA Victor to my knowledge never made an integrated amplifier that included a high-gain magnetic phono preamp stage on the same chassis with a power amplifier. If you try your grounding procedure on a high quality Fisher or H.H. Scott integrated stereo amplifier it may not work out so well. We're trying to help you do better work. Even an RCA Victor amplifier will benefit from it.

Filter capacitors should be grounded as close to the power transformer H.V. winding center tap connection as possible in order to minimize ground currents through the chassis.

-David

So I was intrigued and did a couple of little experiments. I connected my daughter's Fender Squire Strat to my oscilloscope, gave it a couple of strums and was surprised to see how high the output was -- 800 mV peak-to-peak. That's 280 mV RMS, close to the level of some of the better modern ceramic cartridges. Let's compare this to input sources that other amplifiers have to deal with:

Moving magnet, moving iron or variable reluctance magnetic phono cartridge: 5-8 mV typical
Moving coil magnetic phono cartridge: 0.5-1 mV typical
Low-impedance dynamic microphone: 1-2 mV typical.

So a guitar will easily saturate an amplifier with reasonably high but not exceptional gain.

Which is what they seem to have. My wife has an Ampeg re-issue SJ-12T Superjet amplifier. I plugged an iPod into the guitar input, set the iPod at about 50%, and the amplifier level at about 40%. The volume was normal listening level at this point. The iPod level was what would be used for a PA amplifier with a high-impedance microphone input, as well as for the microphone input to my Newcomb TR-1625M which I use with the iPod a lot. So I would say that a guitar amplifier has about the same gain as a typical PA amplifier.

I can do a formal sensitivity measurement on the Ampeg, but not for a couple of weeks.

Don't be fooled by the number of 12AX7s. Not all of them are providing straight gain. Guitar amplifiers have a number of effects, including tremolo and reverb, which require 12AX7s. Some of them also have multiple input channels, including clean and distorted, which require separate 12AX7s. So the number of in-line straight gain stages may not be that high even if there are eight 12AX7s on the chassis.

When comparing a guitar amplifier to a hi-fi amplifier, it's important to distinguish between the different kinds of hi-fi amplifier. The gain of an RCA Victor amplifier with no preamp, high-level input only, will be much lower than the gain of a hi-fi integrated amplifier with a phono preamp or a PA amplifier with mic preamps. Apples and oranges.

-David

David, don't worry I am not fooled...LOL Please give me enough credit that I know the difference between a 12AX7 being used as pre-amp tube and one used for the tremolo or reverb circuit or run in separate channels. I may be very humble about my knowledge but don't mistaken my being humble for stupidity. There are indeed several high gain guitar amps I know of off hand that use 4 or more 12AX7s for up to eight stages added gain in a channel. I can supply models and schematics for proof but I don't think that was the problem here.

You hit the nail on the head with your last statement. I was talking apples and you oranges. I was comparing a guitar amp to consumer grade vacuum tube hi-fi or record player amps which is what MagicBrain started this thread about and what is the main topic of discussion 99% of the time on these phonograph forums.

T