Shotgun renovation of Fisher 30-A

Why? ...why not


I picked up a previously modified Fisher 30-A last year, and meant to use it as a good training piece for my fledgling hobby in vintage electronic repair.

I recapped my Fisher FM-50 tuner along with some other minor repairs such as new X/Y caps, thermistor, power cord and a general restoration. That worked out well and looked great. So I decided to give the 30-A a try. You can certainly see a number of my questions related to the 30-A and FM-50 in the last months.

Well, other than some minor nits such as a cord strain relief, the 30-A is finished.... for now (always the possibility for more tweaking).

I've recapped the entire unit, using one of the new CE (Mallory) can electrolytic caps. The original spec was 40/40/40uF 450/350/350V, and the replacement is 40/40/40uF 525V. I do have an outstanding question in another thread if larger filter capacitors might be better for the amplifier though. We'll see what comes of that question and if larger filter caps might be of benefit, especially with regard to hum.

I've added a CL-140 thermistor for a softer start-up. I used this instead of the popular CL-80/90 unit because it is closer to the rated draw of the amplifier. I suppose this was less necessary in the 30-A since it is tube rectified than the FM-50 which is SS rectified, but what the heck

As recommended here, I also added X/Y rated "death caps" on each main leg and across the legs for filtering.

You may notice that I used carbon-comp resistors again. The originals were actually in good shape and most tested within spec. So again, no need to replace, but this was more of a learning project for me than anything else. I replaced all of the original 1/2W CC resistors with 2W CC resistors, all are now 5% spec and all except two are mil-spec. My limited research suggested that larger wattage CC resistors would likely be more stable and produce less noise than the smaller wattage units given their ability to be less susceptible to heat induced performance changes. There was also some suggestions that the larger diameter 1 and 2W CC resistors performed better generally. I'm of course not sure that it completely true, but thought it would be an entertaining exercise. The real issue, and usually reason, that the smaller 1/2 and 1/4 W resistors are used was due to physical size. As you can see, the 2W resistors were a bit of a challenge to fit due to size.

I replaced the coupling and input capacitors with Angela (Jensen) paper/oil capacitors. The one that is perpendicular to the other two still has a bit too much play/wiggle due to the long leads than I'd like. I don't think there is any real chance of it "settling" and touching anything or vibrating excessively, but I'll likely tie it down a little more securely.

The one electrolytic downstream is now an Elna Silmic II with the same values as spec. The voltage on this part of the circuit is supposed to be about 10V, so I think 25V is good. I was unable to find a 35V or 50V handy, but may replace that later as doubling the spec voltage on this was a recommendation I had received for longevity of the cap.

I left the one disc ceramic capacitor in place, which I think is an RF ground filter (C2) and should be fine. I may replace it later if my hum issues continue, more on that later. The other two capacitors (C4 & C5) are cylindrical ceramic axial units that should also be fine. I've heard they can have special thermal specifications that are hard to replace. But I may still take a look at these three if it seems recommended. I'm trying to understand that part of the circuit, and wonder if that is for NFB?

The other three wirewound resistors were original the typical cement brick type, and they've been replaced with Dale and Ohmit WWs of the same resistance and a little more headroom on wattage in a couple cases.


So in the end, does it still work? Good news is yes, no magic smoke was released :yes:

Following the advice here, I painstakingly removed only portions/sections of the old components at a time, and carefully (3 times) noted the new components and their connections on the schematics and digital photos.

I checked the main DC voltage and per spec it should be:

315V, actual measured 326V
270V, actual measured 302V
385V, actual measured 388V

The only voltage over 10% is the 270V going to the 12AX7.

Any thoughts on if these are OK, or if some adjustment in resistor value is needed?


I suppose this is due to variances in the components and higher line voltages today.




So, now complete does it sound better, I can't really tell! *laugh*

....but it sounds "good"


Again, I didn't really expect some sonic leap, and it was more about the fun of reworking the parts and getting some practice.


Now that all being said, I still have the same light "hum" that it had before the rework. It definitely is noise getting into the input stage, as I can short the input and it is dead quiet. And at least part of the noise is coming from my house/environment because the amplifier is noticeably quieter if plugged into an A/C circuit upstairs than several in the basement.

Again, all that is left in the amplifier that is original are the three ceramic capacitors, transformers, potentiometer(cleaned and quiet) and of course the chassis, sockets, fuse holder, etc.

So if anyone has any thoughts on the source of the hum based on everything above and the photos following, please feel free.

The hum is manageable, but it would be nice to see if I can eliminate it. I have a 7025 tube coming to test as well to see if its spiral filament may help over the 12AX7. And I've tried several different rectifier and 12AX7 tubes with no difference in hum.


With all that being said, it has been a lot of fun to take this apart and reassemble. I think it will be a good starting point for more complex projects and beginning to understand how these circuits work versus simple part-for-part replacement, the latter which is where I mostly am at this point.

Thanks for all the help over the last months. And if anyone has any thoughts on those final three ceramic capacitors (to keep or replace with something else) and the hum issue, please feel free to comment!

Thanks again,
Al

Fisher 30-A Schematics and Parts List

Here is a thread where the schematics for all three Fisher 30-A versions are shown, and the accompanying parts list.

The version in my case is the "10000 series" which is the 2.5Ohm output.


http://www.audiokarma.org/forums/showthread.php?t=476973

Post renovation photos.

Some photos post-renovation.

...it still needs a cord strain relief. The original was too tight for the new cord, so I've still got to fit a new one of the appropriate size.

Hum shoots through roof if I touch input capacitor

I don't know if this is a clue, but when I bring my finger close to the Angela capacitor between the input jack and pot, the hum gets much louder and is very loud when touched. This is true regardless of the orientation of the capacitor (outer versus inner foil) and also was true when I had the Illinois poly capacitor in the Angela's place.

Based upon some other reading, and the "finger test" I'm not sure if this is really a meaningful observation and may be completely expected.

But then again, perhaps not. So I thought I'd ask if this seem unusual, and if so what the solution might be if it is this part of the circuit that is significantly contributing to the hum.

Thanks,
Al

Al -- with the type of hum you describe, you could eliminate most of the hum by:

1. Making sure a bottom plate is attached, and

2. Reconfiguring the heater circuit ground so that it is a balanced system. That is, the heater supply wiring should be tightly twisted wires going from the transformer to the first tube, and then from tube to tube, and the circuit should then be grounded with two 220 ohm resistors -- one from each side of the heater supply to ground. The old ground connection from one side of the heater supply should of course be removed.

With these two items in place, you'll have your hum on the run.

I hope this helps!

Dave

dcgillespie said:

Al -- with the type of hum you describe, you could eliminate most of the hum by:

1. Making sure a bottom plate is attached, and

2. Reconfiguring the heater circuit ground so that it is a balanced system. That is, the heater supply wiring should be tightly twisted wires going from the transformer to the first tube, and then from tube to tube, and the circuit should then be grounded with two 220 ohm resistors -- one from each side of the heater supply to ground. The old ground connection from one side of the heater supply should of course be removed.

With these two items in place, you'll have your hum on the run.

I hope this helps!

Dave

Click to expand...

Hi Dave,

Thanks for the additional tips.

1) The 30-A didn't come with a bottom plate, but I can certainly fashion one, with perforations for ventilation of course. I assume this would be to complete all fours sides of shielding as the bottom is open in the current configuration.

2) This sounds like a fun modification to attempt. So let explain what I see today with the current configuration and recount what I think you are suggesting.

As you suggest, the current configuration already uses two tightly twisted wires running from the power transformer (green/yellow) to the four tubes. The two heater wires that run throughout the amplifier are brown and black. The brown wire is attached to the green transformer output, and the black wire is attached to the yellow.

The EZ81 rectifier tube is first in the chain tied to the green/yellow of the transformer then chained through to the 12AX7, and this is where the heater circuit is then directly tied to ground via the black wire from pin 9 on the 12AX7. This ground is direct to chassis with no resistor inline.

The two EL84 power output tubes' heater circuits are tied directly to the green/yellow transformer and then run along the outside edge of the chassis as a tightly twisted pair and then chain from one EL84 to the other as you suggest.

But as you observe, the only place this circuit is grounded is at one point, off of pin 9 on the 12AX7.

So I wonder a couple things.

A) I wonder if hum is being introduced into the 12AX7 through its ground connection for its heater as where it is grounded is a tie for several other components such as R8 and R10 (I believe from a quick glance) and the potentiometer. I wonder if moving the ground from the circuit somewhere else, even if unbalanced, would help.

B) But if I follow your lead, which I'm really more interested in doing. I may need some help understanding exactly how to do this.

Clearly I would remove the current grounding of the black wire via pin 9 of the 12AX7 to a lug on the terminal strip nearby.

But there are some complications/questions here. If you can see in my photos, pin 9 is not only attached to ground, but it also has a tie to the center post of the 12AX7 socket. I am not sure why as nothing connects to that center pin socket. Would disconnecting the ground via pin 9 have any impact to this center socket connection? I wouldn't think so, but thought it good to ask.

Also, the ceramic 100pf N1500 capacitor C2 bridges between pins 6 and 9 of the 12AX7. C2 is supposed to go to ground from pin 6, so would it then be correct to disconnect the leg of C2 from pin 9 and connect it directly to ground?

So at this point what we're left with is:

1) 12AX7 pin 9 no longer grounded to the terminal strip lug.
2) 12AX7 pin 9 still connected to the 12AX7 center pin socket, no effect, but now ungrounded.
3) C2 disconnected from pin 9 and reconnected to ground lug on terminal strip directly.

Now the heater circuit is ungrounded completely, and C2 still has a ground.


So then your suggestion is to ground both sides of the heater circuit with 220 Ohm resistors.

Is this as easy as it sounds? It sounds like I simply need to ground the green/brown wire to chassis with a single 220 Ohm resistor, and similarly ground the yellow/black wire to chassis with a single 220 Ohm resistor?

Does the location where the resistors are tied in matter? For example could I put these resistors right where the green/yellow wires mate to the black/brown wires at their terminal strip in the following photos (see note in red with arrow)? The terminal strip there would make that very simple. If that location isn't a good idea, where would be better?

Also, what wattage should these resistors be?

I think I've got the general idea, but I just want to be sure.

If you don't mind, I'd also like to understand why the choice of the 220 Ohm resistor value when the original single grounding had no resistor at all. And doing some reading on similar setups, I've more often seen 100 Ohm resistors specified. Why would the 220 Ohm be a better choice in this amplifier?

Thanks for the great suggestions!

Al


P.S.

In the attached photo with the black oval drawn, one can see the tag strip where the transformer hands off to the two heater runs, and where I was thinking it might be ideal from a convenience standpoint to put the two 200 Ohm resistors and tie them to ground right there.

Al -- You understand the concept perfectly. However, let me suggest:

1. While pin #9 of the 12AX7 should no longer be grounded, the center post of the tube socket should still remain grounded to the terminal strip ground lug. This helps to provide a ground plane within the wiring of the tube socket which works to minimize hum. C2 can then ground at either directly at the ground terminal or the grounded center post of the socket.

2. It would be best to connect the 220 ohm resistors at the rectifier tube -- or if not there, at the output tubes. The point is to get them away from the 12AX7 area. The wiring at that tube is the most sensitive portion of the circuit signal wise, and these resistors -- like all wires and components -- will radiate energy from the current passing through them -- in this case, 60 Hz energy from the heater circuit, so it is best to keep them away from the input area of the amplifier.

3. The value of the resistors is hardly critical, but chosen to: (A) Not place too much of the load on the heater winding of the power transformer. The two resistors are effectively wired in series across the winding, and so they do pull some current -- but with the value chosen, not enough to be of any concern, and (B) Be small enough to still represent a low impedance path to ground for the heater system -- this to keep hum and noise to a minimum.

Finally, Fisher commonly used 220 ohm resistors in this application in many of their units, so this value is in keeping with that which Fisher engineers used. In reality however, any value between 100 ohm and 470 ohms (each) would work just fine.

I have little doubt that by performing this mod and fashioning a proper bottom plate, that your hum will be a thing of the past.

Good luck with it!

Dave

Thanks Dave, sounds like a plan.

One last question: What wattage is recommended for these resistors, and would a particular type be better suited, e.g. wire-wound, carbon comp/film, metal oxide/film, or ?

Thanks again,
Al

Carbon film 1/2 watt is fine.

I notice from your rebuild pics (which looks great!), that you have used almost exclusively (what appears to be) 2 watt resistors throughout the unit. While this is great for stability and noise generated from the resistors themselves, it also makes them a much bigger antenna to pick up noise as well, so there are trade offs to be had when using that approach. Bigger is not always better, with -- as always -- the answer being in the middle. It's not something you need to do anything about, but it is something you need to be aware of for future projects.

Dave

Thanks again, will do. Good advice on the 2W resistors. Hopefully the additional shielding from a bottom plate will help with the "antenna" potential to some degree as well. The 2W size were definitely a challenge in and around that 12AX7!


Another general question for anyone following this thread.

You will notice that there are three original capacitors left in the amplifier.

C2 which I think is an RF ground filter and should be fine being a quality ceramic disk (although I currently have no way to test it), and it is hard to replace/find anyway given its specific N1500 specification. So I plan to keep it.

The other two capacitors (C4 & C5) are cylindrical ceramic axial units of very small values that should also be fine technically since ceramics rarely go bad. But given that they are part of the link to the speaker output, I wonder if something like some silver mica capacitors might be a slight upgrade? Thoughts, or just leave them be?

Thanks,
Al

The purpose of C2 is to tailor the response of the design after the NFB loop is closed. Without it, the amplifier would likely become unstable with certain loads attached.

While it does certainly lower the overall response of the amplifier to well below the lower RF regions, that is not the intended purpose for its presence in the design. The effect of NFB is to try and extend frequency response to an infinite high frequency value, which of course never happens, but does mean that it can reach as high as 100 kHz or more in practice with the better output transformers when the NFB loop is closed. C2 restricts this response to a more practical 40 kHz or so to prevent instability under certain load conditions.

C4 & C5 are part of the NFB circuit itself, tailoring the high frequency response of the NFB signal, with specific values chosen to match the high frequency characteristics of the OPT used to those of the basic amplification block design. Once determined, then C2 is used to produce the overall response characteristic sought.

Replacing these components with silver mica components is more of an effort in ensuring value accuracy rather than so much producing any specific sonic signature.

Dave

Very interesting. I suspected some of these may be involved in the NFB circuit, but it sounds like I need to do some more reading on that subject to understand how that works.

So, if I had a 47pf 500V 5% (versus installed 10%) silver mica on hand for C4, and a similar 150pf 500V 5% (versus installed 10%) silver mica on hand for C5, would it be attractive to experiment with replacing these? Or is there really no benefit to replacing the existing cylindrical ceramics?

C4 is made by Erie, and is a solid brown cylinder. I can't see who made C5, but it white with red ends.

C2 is made by RMC, which seems to be pretty ubiquitous throughout many electronics of this era, especially Fisher. Along with disk ceramics being pretty indestructible (within reason) and that this value with N1500 specs isn't terribly common, I don't plan to mess with C2 as I'd be surprised if it needed replacement.

Thanks again, this is exactly why I decided to tear completely into this 30-A, to force answers to some of these questions and learn more about how all this works together.

BTW, did you suggest the carbon films for any technical reason, or simply because they are low cost, reliable, and will "work"? I wasn't sure if was tied to any issues like inductance with wire-wounds and some metal films/oxides.

Also, can you take another look at the photos I attached above? I've attached a fourth photo, and "P.S." where I clarify where I was thinking about mounting these two resistors. This new fourth photo includes a black oval circling the area where I was thinking it would be very convenient to mount the two 220 Ohm resistors, and they would be far away from all the tubes. I could connect the two 220 Ohm resistors at that area on the terminal stip on the top two lugs where the transformer hands off to the brown/black heater wires, and the third lug is grounded and is where my three X/Y bypass caps are also grounded.

Thanks again,
Al

Your location for the resistors is perfect.

Carbon film resistors are perfect for most audio work. Many feel they can hear the difference between the different compositions used in various resistor products, but double blind testing has shown time and again that they would truly have to have golden ears to hear such differences. On the other hand, wire wound resistors should only be used in power supply work where possible.

Again, I would only replace C4 and C5 if you have reason to suspect that the originals are off value. If you do replace, them I would certainly use the original values when doing so.

Dave

On the other hand, wire wound resistors should only be used in power supply work where possible.

Click to expand...

Interesting, do I understand you to mean:

a) Wire wound resistors shouldn't be used anywhere except in power supplies.

and/or

b) Power supply resistors should always be wire wound.


(I am curious as I do have a 22k Ohm carbon comp 2W resistor on the last power supply filter on my 40/40/40uf can. It was originally a 1/2W carbon comp, but I upsized it to 2W. Would this be better served by a wire wound unit?)

.... and I guess this leads to the inevitable "why"


Inductance? Safety?


Thanks,
Al

Better understood to mean, use wire wound resistors only where carbon or metal film units cannot be had in the necessary wattage rating.

Dave

Hum got louder....

Dave,

1) I removed the ground from pin 9 from the 12AX7
2) I reconnected the center socket of the 12AX7 to ground
3) I reconnected the C2 capacitor to the center socket, hence ground. This ground lug on a terminal strip is shared with R10 and R8 and the ground wire from the potentiometer.
4) At this point the heater circuits are ungrounded.
5) I connected a 220 Ohm resistor from one leg of the black heater circuit to ground, which is shared on a lug with two of the X/Y capacitors from each leg of the A/C main.
6) I connected a 220 Ohm resistor from one leg of the brown heater circuit to ground, which is shared on a lug with two of the X/Y capacitors from each leg of the A/C main.



Unfortunately when all done, the hum was now quite a bit louder than before, perhaps twice as loud. It was also very sensitive to if I touched the case, where it would then get a bit quieter.

When I attached my FM-50 tuner, the hum dropped to about the level it had been before when the tuner was attached, which is tolerable.

But unattached, it is much louder now.

Again, as before, if I short the input it is dead quiet.

I still don't have a bottom plate yet, but I don't think that is the primary issue.

Any ideas based on these observations?

I have attached a photo of the installed resistors.

Al

One thing to check is the integrity of the ground you are using for your heater circuit. Based on the SA-100 rebuild I just completed, the mechanical grounds in some of the older Fisher chassis deteriorate with age quite significantly. In the particular unit I was working on, one grounded terminal on one strip read over 15K ohms to ground -- this due to oxidation with age. Use your ohm meter to to check the resistance from the grounded terminal to a known good chassis ground point. There should be no discernible resistance over that which your meter displays with the probes shorted. I would check all the ground points that are grounded by a mechanical connection.

Let us know!

Dave

Thanks Dave, as requested I've checked the resistance.

The 30-A has four ground points.

I tested by clipping one lead to the far end of the chassis and using a spring test hook on the other end of the DMM. The range was set to 200 Ohm


Going left to right, starting at the power supply:

1) Terminal strip under the power transformer, four terminals, 2nd lug from top is grounded via a rivet. This strip is where the transformer hands off to the heater wires on lugs 3 and 4 and is the photo where I've added the two 220 Ohm resistors. The rivet on this strip has a paper insulator inside the case, but is indeed grounded on top via the rivet. However I had added a green wire to strap it to the chassis for a good ground. You can see that wire in the photos. I tested with the wire connected and disconnected. Note, that connected or disconnected, this green extra ground didn't make any impact to the hum. The resistance tested from both resistors through the ground lug, with and without the green wire connected, was .3 Ohm

2) The second ground point is where the can capacitor and red/yellow center tap ground from the transformer are bonded together to the chassis. You can see this in the photo. All three twist lock locations test .4 Ohm

3) The third ground point is the strip near the 12AX7. It is the second lug from top. As noted this ground point shared R8, R10 and the ground for the potentiometer. With this latest modification, the 12AX7 is only grounded to this lug via its center post/socket and from pin 6 via C2. It is also riveted, but I had soldered it to the chassis. This lug tests .4 Ohm

4) The third ground point is the strip beside the EL84 output tubes. This lug has several components tied to it for ground, which I believe includes R4, R11, R5 and C3. The speaker output common is also grounded to this lug. It is also riveted, but I had soldered it to the chassis. This lug tests .4 Ohm.

The "hum" seems to have two components. When I am not touching the chassis, the "hum" is mostly a buzz. When I touch the chassis, the "hum" seems to become two components, a buzz and hiss. The hiss is very very low, hard to hear, but there. The buzz is the major component and drops by a significant amount when I touch the chassis.

Again, when the input is attached to my FM-50, the "hum" decreases back to the level it was (or seems) before this 220 Ohm modification. It is a tolerable amount, but then again seems abnormal and is a bit annoying with the high sensitivity speakers I've got.

It wouldn't bother me too much with the FM-50 attached, except that some folks here and on some other forums have told me that their 30-A amplifiers were "dead quiet". I also don't feel too bad because it hummed when I got it, so I don't think I did anything "new" to it with my work. But I do think there is something not right. What's funny is that I bet it is something simple, and once found will be a "DUH" moment *laugh*

I'm going to fashion a temporary bottom plate out of cardboard and foil to see if that knocks any of the hum down in the meantime. I'll let you know how that goes.

Thanks for all the help. I appreciate the time you're giving to this problem!

Al

Hi Al -- As ol' Sherlock would say, this is becoming a two pipe problem!

A couple of quick thoughts to chase down:

1. Does the power transformer on this unit employ a static shield? This is that often seen large copper strap running the thickness of the lam stack on two sides of the power transformer. If you yours doesn't have one, there's not a lot that can be done, but when one is present, it sure is a lot easier to chase hum out of the chassis from the power transformer. Sometimes the strap (when present) is painted black making it harder to see.

2. My attention is now focusing on your AC wiring/grounding scheme. It appears that you are still using a standard two wire power cord. If so -- coupled with the multiple AC caps you have installed -- this could very well be part of the problem. Further if so, it is also part of that larger issue where in an effort to be super safe, it caused more problems than it solved. That can happen.

It is fine to have a cap across the line to filter out noise. It is also fine to have one cap to ground from one side of the line. Notice that was the way the original design was. But with a cap from both sides of the line to the chassis, in tandem with only using a two wire AC line, then a voltage divider is effectively formed which is trying to place the chassis at 1/2 the AC line potential. That produces a potential hum problem waiting to happen.

If you want a cap from both sides of the line to ground, then a three wire AC line cord will likely be needed to knock the hum down. On the other hand, using a three wire AC line cord will likely cause ground loop hum when the amplifier is used in conjunction with other components that also employ a three wire AC line cord system.

To eliminate these problems, my approach has always been:

1. To always use an AC cap from one side of the AC line to ground with power transformers that do not employ a static shield. For transformers with the strap, you can often get away with no cap at all.

2. When a cap is used, it is always connected from just the neutral side of the line to the chassis.

3. A two wire polarized line cord set is then used to ensure that the unit can only be plugged in one way, so that the single cap is always connected between the neutral (large prong) side of the line and the chassis.

Using this approach, the chassis is then always referencing the low side of the line, producing the lowest hum. With your system of having both sides of the line referencing the chassis, and no independent ground to lock the chassis at ground level, it may be no wonder then that this sucker is so sensitive to the touch.

The rework of the heater system was hardly for naught. Using a balanced system and eliminating the heater current from flowing through the audio ground (chassis) is always a good thing. It is likely that any improvements that effort made however were simply drowned out by the AC wiring issue.

I hope this is it. Let us know!

Dave

Dave,

Great suggestions.

To answer your question about the transformers, neither have a shield around the laminations.

I'll definitely take your recommendations on the AC filters into account. I don't have a polarized plug on the new plug wire, but I can certainly mark the neutral per your suggestion. Removing one of the filter caps is certainly simple.

Question for you though, looking at the schematic and photos of the current wiring, which leg should I choose to be neutral in your opinion?


But here's the final bit of news.... the temporary foil/cardboard bottom just about completely eliminated the hum/buzz. I feel silly, as this of course was the simplest thing to mock up and test. I'll be honest that I really didn't think that would be the problem as all 30-A amplifiers have an open bottom. But perhaps my 30-A is just more sensitive and/or my home has more RF noise. It is certainly possible the latter is true.

Thanks again for all the help!
Al