• 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

HF-81 (and general) grounding question

lenos

lenos

New Member
I recently completed a ground up rebuild of an HF-81 using all new components except transformers, switches etc.
It took a bit of debugging to get it all going but then I had a few moments where I could hear what this thing might be capable of before other issues popped up.

One of the things I'm struggling with is hum.

From reading many other posts it seems that the general consensus is that most hum issues are a result of some sort of grounding issue. To be honest I didn't follow the directions exactly regarding grounding when I put this thing together, and so am revisiting that. In doing so it seems to me that the path to ground is somewhat circuitous in lots of places.

In looking at the intended wiring and schematic it looks like all the cathode resistors and cathode bypass capacitors eventually end up being grounded through what is referred to as Grounding Lug "G". I'm sure there is a logic to this (which I ignored). The thing that bothers me is that there are a lot of intermediate connections. An extreme example is for R43 which connects the cathode of V5A to ground. It connects from V5-8 to Terminal Board (TB) 7-1. From there the route is TB8-2, TB21-1, TB11-3 and then Grounding Lug "G". It would seem to be a lot simpler to connect a ground wire directly from TB7-1 to Ground Lug "G" and do the same for all the other intermediate ground connection points. I would plan on replacing Grounding Lug "G" with something having more room for connections.

Is there anything conceptually wrong with that?

My thought is that if I can get away with two soldered connections rather than eight (for this particular example) there is less potential for things to go sideways. Also, some of those terminals get mighty crowded when you stuff them with shields and multiple interconnect wires.

I recognize that grounding is a lot like voodoo but would like to hear other's thoughts.

Thanks.
 
Register to hide this ad
I am hoping your HF-81 is now humless. Even though the HF-81 uses AC filament heaters, these amps can have minimal noise.

Essentially with multiple grounding points, the steel chassis itself is relied upon.

Fortunately, the terminals which are grounded are not installed with rivets. Rivets were usually brass or aluminum, causing eventual electrolysis with steel chassis and high resistance junctions. So, in any very vintage amp with multi-grounding points, sometimes the grounding points need "augmented grounding" from a nearby, lower resistance (to chassis ground) ground terminal.

Using a sensitive VOM, check each grounding terminal. Any which are more than .7 Ohms, might need a wire added to a nearby lower resistance grounding terminal.

In my Eico HF-85 preamp, using the same preamp arrangement as in the HF-81, I found the lowest resistance grounding point near the phono stage and ran a ground bus wire from that point, then along the selector switch, in order to assure lower hum than stock. That worked a charm, as the high gain stages would be the most likely to amplify any noise.

If that does not help, my guess is the hum can also come from a poor hum control pot or filament wiring too close to the power switch. Of course, power supply caps are the most obvious culprit for hum probs.

My pet peeve with this fave amp is the 6CA4/EZ-81 wiring. The diode sections were designed with each rect. tube as a half wave rectifier. In time, as one EZ81 weakens, unbalance causes hum/noise and possibly even damages the high voltage winding. Rewiring as parallel, full wave rectifiers solves this issue.

Any HF81 restorer should check/search this forum's threads, as many of the HF-81 issues have been discussed.
 
Thanks for the reply.

Power supply caps are new so that shouldn't be a problem.
I have found that the hum is both 120 and 60 Hz.
The power supply itself hums pretty loudly. I have a new one on order.
I am getting pretty frustrated with this thing, it's not getting any better. I am compiling a list of things I've tried and the results and when finished will post hoping others might benefit.
Regards,
 
Thanks for the info. That document seems like a good reference. I'm on another project that is just about finished and after that will attack the hum (if the current project doesn't crater). I have a new power supply coming next week so it will be interesting to see if that has any effect.
I'm basically a mechanical type of guy and never trusted electronics because of the seemingly random rogue electrons etc. This is a learning experience for me.
 
Lenos, do you have any pictures of your HF-81 and what you’ve done? I have an 81 as well. Some similar issues to you also.
 
One important thing (maybe the MOST important!). The power supply return (center tap of the transformer) MUST go directly to the negative terminal of the first filter cap - not to chassis, not to a bus, DIRECT! If you have to, splice a wire to the center tap wire so that it will reach. That point can now be connected to ground. The first capacitor has fairly high ripple current, and that current should not be flowing in the chassis.
 
Tube Active, thank you for pointing out that the rectifier tubes are wired individually as half-wave rectifiers! But, while it's conceivable that one tube could wear out sooner than the other potentially causing a hum problem, in my experience, rectifier tubes don't lose emission over their lifespan. They seem to test nominal until the day they either go open or short. And the most-common failure mode of a dual rectifier tube is one half goes open, hum increases greatly, and then the other half goes open, and then you have no sound at all.
 
Hi audiocarp, I rewired my HF-81 rectifiers into full wave rectifiers in parallel, after I had that exact, hum and reduced output, problem occur. While your experience may differ, have you heard how many HF-81 power trannies have been replaced thru the years ? I suggest the reason for the many failures might have been the "mis-wired" EZ81/6CA4 tubes...as well as a slightly inadequately beefy, power tranny...

With these low power amps, disasters can be avoided more easily. If you research the 1940s and 1950s, higher power amps using dual 5U4s or 5V4s, you might be surprised how many also use the dual half wave rectifier approach. This engineering design was implemented before decades of usage could prove eminent destruction...Has it ever been written that similar tubes should wear out similarly or is this just a hopeful request ? Parallel full wave rectifiers would be a safety measure, keeping the amps living longer, IMHO...
 
Tubeactive-

Excuse the long reply; it's been a little crazy this week...

Let me make sure I comprehend what you did: You wired one of each 6CA4's plate to one end of the high voltage winding and the other plate to the other high voltage winding, so that each tube is acting like a full-wave rectifier? That's interesting. But, in terms of loading, the rectifiers are seeing the same current, so I'm not seeing how this is a better design?

I did a little research to see what the experts had to say and on p. 1165 of the Radiotron Designer's Handbook, in the Power Supply chapter under Amplifiers:—, it says, "Parallel operation of similar types of vacuum rectifiers is possible but it is preferable to connect together the two sections of a single full wave rectifier and to use a second similar valve as the other half-rectifier if full-wave rectification is required." And on p. 221 of The Radio Amateur's Handbook, 1958 edition, under Operation of Rectifiers says, "Rectifiers may be connected in parallel for current higher than the rated current of a single unit. This includes the use of the sections of a double diode for this purpose." Neither reference elaborates on why this is better; it may have to do with each tube seeing a smaller voltage swing (from half of the winding to the center tap)?

I think the transformer failure has more to do with EICO under-rating it and using an over-rated 3-amp fuse and the users foolishly using these amps with leaky filter caps well past their lifespan?

Don't know if you're being rhetorical about 'similar tubes should wear out similarly'? I never said that. I was speaking only of dual rectifiers which will test almost identically from one plate to the other plate when new, and measure the same after being used for many hours. And there is no variance from the bogey value as is always seen with a random lot of new triodes and pentodes.

Finally, I take issue that designers in the '40s didn't know how to design power supplies. If we acknowledge that the first commercial hi-fi amp was introduced in 1947, which means the radio tube industry was already well over 25 years old when they started paralleling rectifier tubes in amps. I think the failure mode you're suggesting would have been discovered before then.

I wonder why EICO didn't use the 6BY5 instead of two 6CA4s? But, that's a much taller octal tube; I'm guessing it wouldn't fit under the cage?
 
The lit you cited is worthy....Theory is an important guide. Engineering practices can supersede written text.

When you have tested and tried enough old, well used rectifiers, you will be surprised how many still work. But, well used, dual rectifier tube sections do not typically stay matched over time.

As I said earlier, we have different experiences. In the 1980s, when I had numerous dual 5U4, 5AW4, 5V4 etc. rect. tube equipped amps, if they were wired as strapped half-wave rects, I quickly rewired for dual parallel topology, simply in the interest of longevity.

Well written theory usually does not take into account all of our long term usages, methods and experiences. I collect and use audio opt and driver tubes, as well as GLobe rectifiers from the 1920s and 30s, often. "Matching" them can require re-testing more often than I wish.

Getting back to the Eico HF-81 rectifier arrangement, I simply decide on the side of reliability. BTW, do you also see the indirect cathode is hard-wired to the filament, thus deactivating the slower warmup ? A few of their amps had that same rectifier circuitry....

Concerning the birth of home HiFi, the desire for better sound reproduction has existed since early radio days. If you bought a radio set, perhaps a swap of interstage or output transformer, or an upgraded speaker, could improve your early HiFi. Certain transformer manufacturers offered wide bandwidth trannies since 1928. Most tranny companies offered a kit of trans n parts to build ur own amps. I believe the tranny companies started our hifi trend. Once electrodynamics speakers possessed usable suspensions, also circa 1928, high fidelity began humbly. Very hifi Phono cartridges are the real "hold-out" in era. I credit the 1946-1947 introduction of GE's Variable Reluctance RPX magnetic cartridges as the enabling factor for every home to have a high fidelity system, post WW-II.

But, perhaps other threads should be started, as the OP intentions are being side-tracked...
 
I’ve been off this project for a few months working on a Scott lc21 LK 150 project with my son. That was going fine until it wasn’t and gave me great conniptions. Thad problem there turned out to be a bad rectifier tube, which was not where I was looking. You buy new tubes and assume they’re good. (Don’t). Anyway with that mostly wrapped up I got back to the Eico figuring that I would need to undo a lot of what I had done and start over from a stock setup. When I put the beast on the bench I noticed that one of the rectifier tubes was white at the top. I went to pull it out and the top of the glass just cane off perfectly as it it had been cut with a glass cutter. Curses! Rectifier tube again, this one had less than two hours on it. I put in a couple of old ones I had in a box, hooked it up, fired it up, and the hum was almost nonexistent. One of the channels is not working correctly so I need to dig into that. I’m sure there must be a lesson in here somewhere, I can’t decide if it’s don’t assume new tubes are good or always check your rectifier tubes first. Probably both.
Looking back on it now, it seems that it makes sense that a bad rectifier tube could give you hum and should be one of the first things to check. It’s part of the power supply which is where the hum comes from.
Now I just need to figure out why the one channel is weak.
 
Tube tester checkers are worthy if u r building the collection. It might be amazing that 60 year old rectifiers can still be found working, but an HF-81 by Eico runs EZ-81/6CA4 in a unique way, prone to humming when one 6CA4 inevitably gets weak. I specified this in my earlier post... You might want to address this, rewiring them into parallel, full wave rectifiers. If someone still ran the amp humming with a weak rect., the power tranny cooks and can go bye bye.

Isn't it surprising a newer tube only lasted low hours before failure ? Long live strong old tubes...

After checking tubes, I would be using a VOM for voltage and resistance checks. A resistor would be suspect unless...Are all the cplg caps already replaced ? If any leak DC Volts to the next stage, bias changes for that next stage...

HF-81s are worth the troubleshooting....Speedy fixes and Best Regards...
 
I built this from a bare chassis with new everything except the transformers, so all components should be good (but that's what I thought about the rectifier tubes too). My immediate task is to figure out the weak channel and then start to re-assess the hum, though when I fired it up the other night it sounded pretty good. I may check back to learn more about wiring the rectifier tubes in parallel. I also have a couple of new PTs that I would like to try at some point.
 
With higher supply voltages, it makes sense to use the dual half-wave connection, as you won't have 800+ VAC aross the tube socket, only half that. Pairing a weak and strong tube will give hum, but I think currents will balance better than if wired in parallel. Mismatch between sections is about as bad in either connection... best to test rectifiers occasionally.
 
You must log in or register to reply here.
Back
Top Bottom