Avery's "First" Maker: The TA-800

[dcgillespie]

Things are progressing smoothly. In the intervening time, with the exception of the volume control, I've finished the lower deck work (based on an inverted position), which in either of the TAs is not a quick job, as the TAs predate Fisher's effort to produce a cleaner layout of the chassis to promote a cleaner build of the unit, and their effort to upgrade their own build skills in general. Let's face it, when the tone controls are on one one side of the unit, but the tone control tube for one channel is on the other side, it results in long wire runs, extra terminal strips, and more crowded wiring in general. This is why in all the later units, the tone/line amplifier stages were moved to the center of the chassis, with each 12AX7 in this section passing both channels. It really cleaned up the build, and minimized different amounts of cabling required between the channels. And while the actual build of the TAs is certainly good, compared to the later models, the solder joints are not as neat, and the dress not as deliberate. If you've ever had a chance to work with a number of the Fisher builds from the mid 50s through to the end of the tube era, the improvement along that timeline is quite apparent.

In any event, with this particular unit, removing the volume control and recapping all of the mylar and small electrolytic caps throughout that level provided not only an opportunity to clean up a previous tech's work around the volume/balance control area, but also some of Fisher's as well! The smaller physical size of today's components helps to clean things up as well. When the new power switch arrives, then the volume control can be reinstalled to complete the work of the original circuits on that deck. The only other work on this deck will be the addition of individual bias/balance controls and potentially some of the EFB circuits as well -- this where the old DC Heater/bias supply's filter can cap once resided.

In yesterday's mail was the CE Distribution order for all the power supply caps, so that timing was perfect. Work on the upper deck (based on an inverted position) can now get started next week. The tubes were mailed out on Friday, so they should be arriving the first of the week, along with the power switch as well. So things are coming together.

A couple of points of note to pass along:

1. The TA-800 uses (in part) .002 uF coupling caps from the final tone stage to the volume control. So does the TA-600. In both units, these caps are rated at 250 vdc. In the 600, because of the slow warming rectifier tubes this is no problem. In the 800 however, the SS power supply produces a start up voltage well in excess of the voltage these caps are rated for. In Rob's unit, one was leaking so bad as to place well over 100 vdc across the balance/volume control combination. With no other intervening coupling caps, advancing the volume control then places a positive high voltage on the control grid of the 7199 driver tube. This not only caused the one channel to be dead (the tube saturated) but also is very likely one of the reasons these tubes tested so poorly in his unit. The other cap was leaking significantly as well, although not nearly to the extent that the one noted above was. Therefore, in at least the TA-800, the voltage of these caps (C114/C116) needs to be raised accordingly. Because of the time constants involved, a 400 volt rating is quite adequate, but 250 volts is regularly exceeded at start up.

2. For all the discussion of what caps should and don't need to be replaced by type, the cathode bypass caps in the tone/line stages (C86, 87,118, 119) were all completely open in this unit. The ceramic looking small electrolytic caps used at this point in time at Fisher for this service have not shown to hold up well. They are not particularly easy to get to, but absolutely need to be replaced with any rebuild of these units.

3. One of the things I've noted with in 800C/1800 design is the poor relative calibration of the signal strength meter when set for the AM band, versus that for the FM band. The TA-800 appears to be much in the same keeping of this observation, if not starting it: the AM section just produces notably less eye tube closure on a strong signal than the FM section does. To be fair, the AM section in this unit has two dead tubes, so I'll know more when the new tubes arrive -- but that's my initial observation anyway. This is not a circuit defect, but a design one. I have a sneaking suspicion that the reason for this has much to do with the same reasons I've stated for Fisher's as designed unusually high threshold for their MPX Stereo receivers to either trigger the Stereo Beacon lamp or produce a steady eye tube in stereo mode with the 400 receiver: noise, and specifically, AM noise. It is the same AM static heard in weak FM MPX stereo broadcasts that you hear in traditional AM broadcasts as well. Therefore, by making only exceptionally strong AM broadcasts produce the same relative eye closure as that of typical FM broadcasts, it will be understood that anything less is an invitation to noise in the reproduced sound. Today however, such calibration is a moot point, with a more practical calibration offering a more usable indication, just as lowering the MPX trigger threshold does. I've already issued a modification for how to address this in the 800C/1800 receivers, but I'll first see how this one does with the new tubes, and then assess from there.

Dave

A few pics to show where things currently are:

BELOW: The right channel tone control circuits, crowded to be sure. The white electrolytic caps can virtually always be considered as bad, as they were in this case. You really can't see it in this pic, but the tubular .002 uF cap at the bottom of the chassis that is under the left tone control was leaking notably due to over voltage at turn on. The brown/dark brown coupling caps of the type used in this unit in my experience rarely leak, but often go intermittent, so they get replaced as well.

BELOW: The right channel tone control circuits recapped. The old selenium rectifier has now be changed out with silicon as well.

BELOW: The left channel tone control circuits recapped, along with the cleaned up balance control wiring. Once the new power switch arrives the volume control can be reinstalled to complete the area. The clip leads lead to an external monitor amp so I can test out the tuner and preamp sections, while the power amplifier section tubes are completely removed from the set. The wiring from the loudness switch has also been replaced awaiting re-installation of the volume control. It has been extended so as to eliminate any potential threat of breakage from physical movement of the volume control, even though the volume control assembly will be more tightly secured before re-installation.

BELOW: The recapped tuner sections.

BELOW: The recapped loudness switch with repaired wiring.

BELOW: For Rob, to help get the ugly image of the blue LEDs out of his head. Note how much brighter the new FM eye tube is on the left side of the set as compared to the existing AM eye tube. The original FM eye tube was much dimmer than the AM tube is now, almost unusable due to being so dim. There is no real reason to have both tuners operating at the same time anymore, so its no real issue, but does show the use level of the AM tube over a new FM eye tube. The set is shown receiving an average strength Atlanta FM station, and a Nashville AM station while I was playing with the AM eye tube closure levels.

 

[gadget73]

Looking good. That FM eye certainly grabs your attention.

I've had mine in simulcast before, its enough to give you a headache.

 

[walyfd]

I switched the AM and FM eye tubes on the 4000 tuner. The AM was considerably brighter indicating a lot more FM use and nor much simulcast use.

All we have is a talk radio station that broadcasts on both bands so I tuned into it. Funny thing is, one band has a slight broadcast delay so I got a bit of an echo effect.

 

[audiodon]

I confess to having avoided these units because of their layout.

 

[nj pheonix]

I was somewhat surprised noone commented on the jammed packed interior. Pretty snug in there!

 

[gadget73]

Is it any more packed than the TA-600? That one is fairly bad too. I have the Sherwood 7000 on the bench at the moment too, I'd forgotten how much of a packed house this thing was inside.

 

[dcgillespie]

Good to hear from you Don!

The TA's are certainly all Fisher, but in retrospect, it is easy to see how hurriedly they were executed in the rush to get them to market -- as compared to later models where their build shows that Fisher was able to catch its breath. The rush is not manifested in terms of reduced performance or capability -- which is as good as any other Fisher product -- but in layout and build execution. The layout of later receivers combined with more polished build skills simply presents itself as a neater, more organized result in their later products. However, I honestly don't hold any of this against Fisher at all, as we have the benefit of hindsight to help in our judgement.

For all intents and purposes, the TA-800 is a TA-600 with SS power supply and bigger output tubes, and a different transformer set of course. Most of the other basic circuitry (which is to say the majority of it) is identical between the two units. Now the 600 was released in '59, which means the drawing board for it was in 1958 -- a period in which Fisher was undergoing tremendous change. Whereas the bulk of the 50s saw Fisher going after the cost-no-object "separates" crowd, the late 50s was a pivotal period of still producing some separates (now as stereo pieces), but rapidly gearing up for Fisher's new vision for the future: integrated stereo amplifiers and stereo receivers, with the idea of separates clearly occupying third chair. And who was to know how strong the demand was to be for the "new" Fisher pieces, and if stereo was even going to be all it was hoped to be? By 1960 however, Fisher's move was already being shown to be brilliant, and sales were quickly ramping up to levels they had never seen before. A new factory was coming into play, new employees and the training that involves, you name it. The point being that in one sense, it's amazing that Fisher pulled it all off as seamlessly, and as well oiled as they did with their 1960 products -- and, the fact of how quickly they got up to speed in improving the build of their future models is no less amazing as well. So, I hardly throw stones at Fisher for the build of the TAs, but simply recognize that they were produced during a period of a rapidly changing Fisher Radio Corporation (and rapidly changing stereo scene as well), which resulted in these particular units being executed with a notably more complex and complicated build. 57 years later, with the benefit of hindsight and an abundance of their future 60s products to hold up for reference, that fact can clearly be seen. But make no mistake about it: the seeds of all the future Fisher vacuum tube stereo receivers were sown in the TA-800. As the title of this thread indicates then, it truly was one of Avery's most significant "firsts" in the ultimate history of the company.

OK. Tubes arrived this afternoon, and the new power switch is due in the next few days. In the mean time, with only the volume control/power switch left to address on the lower deck (based on an inverted position), I've now turned my sights on the upper deck output stage area. Much of the old wiring was removed to allow for a clean installation of:

1. Screen Stability resistors,
2. Cathode current sampling resistors, and
3. Individual bias controls -- capable of working with either the stock bias supply design, or with EFB if possible.

Once again, with the removal of as much of the existing wiring as this required, it gave opportunity to clean up evidence from a previous tech's work, and some of Fisher's as well. Room to mount the new individual bias controls was made on the upper deck, leaving the lower deck area where the old DC Heater/Bias Supply filter can cap was free for the potential installation of EFB circuits. The original output stage area is shown here:

While the new output stage area is shown here:


With the new tubes, I can now start making some basic tests to determine the actual power amplifier performance of the stock design, while waiting on the new power switch to arrive. Those results will then be used to help determine the suitability of adding EFB for enhanced performance. The coupling caps and 7199 screen bypass caps in the left of these pics are all quite good, but will be changed out as well as part of the restoration when work moves over into that area. For those unfamiliar with the underside of the TA-800, the two original Fisher controls seen mounted on the metal plate are the Phase Inverter adjustments.

More when it becomes available.

Dave

 

[gadget73]

Do these use fixed bias, or is it a cathode bias rig closer to what the TA-600 used?

 

[dcgillespie]

No. These are a fully fixed bias design, using the classic DC Heater/Bias Supply format that all the future receivers would use as well.

Dave

 

[derekva]

Retainers link http://audiokarma.org/forums/index....e-tube-retainers-already.573073/#post-7527448
Larry

I'm still amazed with the amount of heat that 7189s at full throttle produce that Fisher didn't invest the $5 in a metal shield between the OPTs and the power tubes in the TA600 a la the 800B.

-D

 

[mrphilco]

 

[larryderouin]

Derek; There is literally NO room to place a flat shield plate btwn the 7189's and the transformers on a TA-600 as is now. You'd have to make one with bends around the tubes and they'd end up touching the shield. If they moved the transformers forward 1/2" it might be possible. But the tubes would be right up against the shield and you could possibly end up with cracked envelopes if they touched the cool shield when hot (thermal shock?). With the retainers on the tubes, they tend NOT to lean anymore so the chances of touching anything is minimized. With the TA-800 either Fisher deepened the chassis a little, moved the transformers fwd a little, or both to get the 7591's to fit in that area.

 

[dcgillespie]

It would be interesting to know if there is any difference in the dimensions of the actual chassis surface between the TA-600 and the TA-800. Within the 800's chassis, space was gained due to the use of SS rectifiers, but lost due to the bigger transformers required -- as relative to the space required for these items on the 600's chassis. For the record, the chassis surface of the TA-800 is 11.75 X 16 inches. Anyone with a TA-600, please chime in!

Dave

 

[RS Steve]

It would be interesting to know if there is any difference in the dimensions of the actual chassis surface between the TA-600 and the TA-800. Within the 800's chassis, space was gained due to the use of SS rectifiers, but lost due to the bigger transformers required -- as relative to the space required for these items on the 600's chassis. For the record, the chassis surface of the TA-800 is 11.75 X 16 inches. Anyone with a TA-600, please chime in!

Dave

Without removing my TA-600 from its cabinet, I can roughly measure 12" front to back for the chassis, so most likely it's the same.

 

[larryderouin]

Mine's boxed up in the shed, at the bottom of the "good junk" stack. (good junk is my wife's term for my working gear. Bad junk is well, bad junk)

 

[Hull Rust]

Mine is a TA-600 Premiere and is 12 3/8 x 16 inches.

 

[dcgillespie]

Ah! So they even made the 800 ever so slightly smaller in spite of the bigger tubes and transformers. Thanks for sharing the info!

Dave

 

[larryderouin]

TA-600 = 10 lbs of _____ in a 3 lb bag
TA-800= 10 lbs of _____ in a 2-1/2 lb bag. Let's see how much more (bigger tubes, larger transformers) we can stuff in this chassis and still make it smaller!!!The bean counters must have loved this exercise by engineering!!!

 

[gadget73]

Depends where I measure my TA-600 from. From the rear edge of the chassis to the front upright panel where the controls mount is 11 3/4". From the rear edge of the chassis to the brackets where the faceplate mounts is 12 1/4".

I suspect these are the same physical size if we measure to the same points.

Isn't the front end the same between the two? If so that avoids a lot of re-engineering effort. In the back, you're ditching a pair of rectifier tubes but adding larger output tubes. Shuffle the input jack box over a bit towards the power transformer and that will give enough room for the larger output stage parts. Probably not as difficult to cram it in as you might expect.

Looking at it, they may not have even changed the spacing on the output tubes. I can see enough space on mine to fit a 7591 bottle in there. As long as the lam stack on the outputs isn't thicker, you can add width and just change the transformer spacing a bit.

 

[dcgillespie]

Gadget -- I was measuring on the bottom side (inverted position) across what represents the sides of the chassis proper -- not including the space between the front of the chassis proper, and the rear of the face plate. That's why I used the term "chassis surface".

Update:

No power switch yet, but progress continues none the less. The driver/phase inverter stages have been recapped, and all resistors checked for tolerance. The resistors were all within a 5% tolerance despite the use of 10% tolerance items. Even the pesky 820K resistors used to power the 7199 screen grids were within 5% of their specified value, which is highly unusual. Usually, this value as well as 680K resistors from this time period drift quite high (as in double the their value), and do so rather quickly over time. These resistors are clearly an improved piece, likely being the last of the carbon comp resistors used before switching over almost entirely to the glass film resistors that are so common in Fisher gear throughout the bulk of the 60's.

Study has also gone into the operating point of the output stage. The TA-800 uses (a measured) 6.5K OPTs, in common with the larger transformers used in the 800B and 800C units. But they operate the screen grids at lower voltages (320 vdc), thus producing less power output. This very likely could have been due to anything from this unit being their first experience with the tube, to the specified tube being a 7591, rather than the A version of the tube. In any event, all of this presents opportunity to coax some more performance out of Fisher's first stereo 800 receiver, so work along that front is starting to ramp up. The OPTs appear to be the same size as those used in the 400 receiver.

Besides installing the power switch and re-installing the volume control assembly, that leaves the power supply sections left to be recapped, and of course, moving towards a go, no-go, and potential installation of EFB. I've frankly kind of held off on full steam ahead with that effort until the volume control assembly can be re-installed, and the volume/balance control tracking and channel balance can be checked and addressed as required. It's just not convenient testing the unit without that assembly installed.

The new tuner tubes made a notable difference in the stability of tuner operation. Originally, the RF tubes in the FM section and the Converter/IF tubes in the AM section were so weak that the local oscillators of these tuner sections wouldn't start until at least 108 vac was applied to the set. With the new tubes, the oscillators now start up with as little as 75 volts applied to the set. As suspected however, the new tubes did not change the relative signal strength indication for the AM tuner, so to have the response of the AM eye tube match that on the FM side, a modification will be recommended similar to that which I published for the same purpose for the 800C/1800 and R-200 models. The re-calibration makes for more useful tuning indications (bringing typical signal strength more within the useful range of the indicator), and also makes it easier to spot weak performance of the AM section as well.

A few pics to document the current work:

BELOW: The original phase inverter section:

BELOW: Close up; As built by Fisher, look how close the red leads (OPT B+) and the terminal they connect to are to the terminal right below -- which is a negative voltage control grid circuit terminal. It is this kind of potentially problematic layout that the revised layout of future receivers would address, eliminating the possibility of a short -- which in this case would instantly ruin an output tube:

BELOW: The finished driver tube area. Note that the terminal strips have much better separation now, keeping a safe distance between the highest B+ potential in the set, and control grid circuit of the inboard Channel B output tube:

BELOW: The complete power amplifier section is all but finished now, pending a slight modification to the output stage screen grid wiring with the potential installation of EFB:


More as it happens.

Dave