I finally bit the bullet. This receiver has been in everyday service since it was overhauled in 1977. It finally fouled out the left-channel output transformer (plate winding shorted to the core). It is a mid-thirty-thousands serial-number unit.

The work completed so far:

Replaced power rectifier diodes with 1N5408's.

Verified B+ cap chains, replacing all softs but leaving those that showed no deterioration (including power factor check). Replaced decoupling resistors along B+ chains (several had drifted off value).

Checked power transformer and remaining output transformer for damage (none found w/meter, full voltage tests pending).

Bypasses into output control grids had been replaced with .1-630v units in 1977 (all paper caps updated at that point). These still check out OK, therefore left alone for time being.

12-ohm quarter-watt cathode resistors installed on output tubes (no 10-ohm resistors on hand).

Added 200-ohm 1-watt screen resistors to output tubes.

Replaced 330K bias feed resistors with 270K's.

Replaced 1K output control grid "swamper" resistors.

Replaced .01uf AC line cap and .01mf cap bypassing rectifier diode to ground. 630v replacements. Also replaced 820K AC-line-to-chassis reference resistor paralleling line cap.

Strip-cleaned chassis and spot-cleaned undersides of 7868 sockets (crud accumulation). Sockets verified OK. Retensioned several of the small sockets with a dental pick.

The bias supply is to be overhauled tomorrow. The original selenium bridge was replaced in 1977 with a Motorola MD980-2 silicon bridge (bridge checks out OK by meter), and voltage divider resistors were updated at that time.

I welcome all suggestions about anything done so far, or anything else that should be done, while I have this unit on the bench. This is my first project of such magnitude since 1992, and you get out of practice over such a time. Spot-and-polish nickel-and-dime repairs on miscellaneous gear, here and there, are not enough to keep a person truly sharp.

Many thanks for all suggestions/commentary, Larry

Thanks for starting this thread. I recently picked up a 400 that was used in a guys garage. I thought I scored it for $80 including the Walnut cabinet.

Unfortunately, a mouse had set up building a nest over the tuner. I am in the process of scrubbing the chassis with cleaner in hopes of removing the crud. The amp works great though. What a sound!!

N9JCQ wrote:



Congratulations on a great find. There are several threads in this section concerning the 400, and also its AM-FM brother, the 800. If you have a renovation in hand, they are all worth investigating.

There are at least 3 different versions of the 400. My unit is a version 2. Fisher went to the second version at serial number 20,000, and a third version at 48,001. What (if anything) happened after that, I don't know.

The major difference between versions 2 and 3 is an extra IF stage in version 3. They also changed the bias bridge rectifier, which was selenium in versions 1 and 2, and silicon thereafter.

Good luck on your project and all the best, Larry

Replacement output transformers installed, and bias supply overhaul finished. Bias voltage divider changed from 1K and 2.2 K resistors to 1.2 K and 2.7 K, with 4K trim pot in parallel with 1.2 K resistor. Tested in site with high-voltage winding floated. The trim pot yields a range of roughly 12 to 28 volts of bias. I plan to set it for 20.

This completes the overhaul except for one thing. I want to install a high-voltage fuse in addition to the line fuse.

If this unit had a traditional full-wave rectifier with a center tap, there would be no question about where to put the fuse (the center tap). In this circuit, however, I am not sure where (if anywhere) a HV fuse can be added.

Can anyone advise? Many thanks, Larry

Bench tests completed tonite.

With line voltage at 120 by Variac and bias at -20, the B+ voltage soars to about 445 before output tubes come up, and then falls back to 425. Combined output plate drain is about 150 ma. All readings remained stable over a 45-minute run.

The B+ voltage does not fall in line with test readings on the schematic, however, until AC supply voltage is down to 110.

Our line voltage here sometimes swings up to 125 nowadays, sometimes down to 115, but never to 110. It usually hovers around 120, and now that I have seen how this set behaves on instruments at such a line voltage, I think I am going to run it on Variac from now on. I hate to tie up a Variac, but it seems the better part of valor.

I would still like to install a high-voltage fuse in this unit, if anyone can suggest how it should be done.

All the best, Larry

Use a bucking transformer rather than a Variac to run it on everyday. A 12 volt 2 amp filament transformer of good quality should be sufficient to do that, and will keep the AC input at about 12 volts less than the actual line voltage. You could also use a wirewound power resistor in the AC input but the bucking transformer operates much cooler.

Increasing the bias very slightly, say to -20.5 or -21 volts at full line voltage, will reduce total current load in the output tubes and make the unit run slightly cooler with no audible difference in sound.

Thank you, Dennis. I think I have a transformer that will work. I'll have to install it outboard in its own chassis box, since there's no good place on the 400 chassis to mount it. But that gives me a good project for the upcoming weekend.

Kindest regards, Larry

The unit is back in service in the synth studio lab, with a two-hour profile run complete. All A-OK, and with 110 volts on the line, nothing heats up like it used to.

(Knock on wood) I think it's a good time to crack a tall cool one, that is, if there's ever a bad time for that . The affair was an ordeal, but it feels good to get back in practice on major work. 17 years away from it was a long time.

I hope this thread can be of help to any other Fisher 400/800 fans who face a major overhaul.

And speaking of that, my 800 is next in line, before it forces me to overhaul it.....

Pros't, Larry

A few final reflections:

The set runs much cooler overall than it did before. Dropping the AC feed to 110V helps, but I am convinced that gearing down the output stage with higher bias and screen dropping resistors helps at least as much. The power transformer scarcely gets warm now, even after runs of 4 or 5 hours, and there is no difference in the sound that I can hear.

Now on to the 800C.

Best regards, Larry

Bump for reference purposes.

Larry

Will yank this unit out of the console over this weekend for photos of the underchassis. I would have done it at the time it was overhauled, but I didn't own a decent camera at the time.

Larry

Fisher 400, top view:



The original filter cap cans are still in place. Two of the cans (the two next to the power transformer) date from this unit's 1977 overhaul, and are still in use. I was unable at the time to get exact-value replacements for the two cans, so I had to add supplementary filter caps underdecks to make up the proper values.

The output transformers came off a carcass 400 I had been holding for years, just in case. A replacement power transformer is still on that carcass (again, just in case! / knock on wood!).

Complete bottom view:



So far, I have had to do nothing significant to the radio section of this unit, apart from replacing a tube or two along the way. Same for the multiplexer unit (upper right-hand corner). I replaced the eye tube in the 1977 overhaul, and it is still in use.

All lytics and mylar/film/paper caps have been replaced. I weeded out the last of them in the 2009 overhaul.

Power supply area:



The new rectifier diodes, safety caps, and several replacement resistors are in plain view. The large "Atom" cap supplements an undersized section on one of the 1977 filter can replacements.

Final filter area:



The filter can in this section was taken out of service in the recent overhaul. The individual lytics replace it. I had to add a terminal strip to mount some of them, along with their associated power resistors.

The notorious Fisher power switch, so prone to failure, is visible in the lower left-hand corner. It happens that this one is still good.

Bias supply area:



The new silicon bridge is mounted on the chassis where the old selenium unit used to be. The two small lytics replace the dual-section filter can that was mounted in the clamp (visible to either side of the white "WP" power resistor). All resistors have been replaced. The bias trim control that I added (a small sealed Allen-Bradley WW pot) is mounted on a wing of the clamp. Its only drawback is that it can't be adjusted from the top of the chassis. You have to go underdecks to get at it.

The replacement silicon bridge is a Motorola MDA-980-2.

Output stage area:



All resistors and caps associated with the output stage are replacements. You can see that those four novar sockets have been hot (you betcha they'll never get anywhere near that hot again!). The 200-ohm screen safeties and 12-ohm cathode safeties are my additions and not part of the original design. Furthermore, the grid bias resistors have been brought down from 330K to 270K. Swamper resistors remain at the original 1K.

I have logged upwards of 200 hours on this unit since October, and (knock on wood again!) so far, so good.

I hope these pictures will help anyone with another 400 to overhaul. It can look daunting underdecks on first impression, but after you zero in on it, section by section, it's very straightforward and easy to follow.

Happy 4th of July to one and all!

Larry

OK Larry, I have to ask.

What is a "Swamper resistor?"

Ron

Hi, Ron. A swamper resistor is a small one (usually 1K or thereabouts) installed in the control grid circuit of an output tube, especially a high-power beam tube or power pentode. The control grid is fed through the swamper.

It's there to interdict (or "swamp") any tendency toward parasitic oscillation of the tube. Higher-power pentodes and beam tubes are susceptible to this. You'll see swampers in series with the output tube control grids in almost any high-powered amp using beam tubes or power pentodes.

Many thanks for the interest and happy 4th of July!

Larry

thanks Larry

Yes, I recall seeing those resistors. Now I know that they have a "special" name.

Have a safe 4th of July.

Ron

And if you check most amplifiers, especially those with 7868 or 7591 outputs, the swamper resisters are usually 1KΩ in series with the grid. Fisher used film resistors and I have yet to find one that was not spot on in value. Very easy to identify...

Great photos, thanks.

Andy

One suggestion-

Put a 10 ohm, 1/2 watt resistor between the output tube cathodes and ground, one per 7868 tube. I just pull out the bare-wire jumpers from the 7868 cathodes and ground, and replace them with 10 ohm resistors. This will allow measuring the cathode current of each tube individually (which will allow you to home in on the right bias supply voltage- typically, I set the most "hungry" tube at no more than 32ma, and let the others do as they will, as long as they're all below that). Also, if a tube shorts, the resistor will "sacrifice itself", usually before an output transformer fails...

Regards,
Gordon.

Absolutely!

Well, guys, I've already got 12-ohm quarter-watt safeties on those cathodes. For ersatz fuses, I think they'll do well enough.

As for determining bias level by reading cathode current, I'll leave that for anyone who's willing to go to so much trouble. Verifying the bias voltage on the bus, and then at the control grid pins, has always worked very nicely for me.



Larry