Ace and Steve’s Big Vintage Summer Adventure:
Update No. 4: I Should Have Known Better
by Steve Graham
If you have been following this series, you’ll know that the original Fisher output transformers were defective. Ace sprung for a pair of new Hammonds that are similar in size and weight to the originals. Not that this means much sonically, except Ace didn’t cheap-out on substandard parts. The Hammond naysayers among the DIY community, to say nothing of the equipment purist, train-spotting types, will likely have built up a good head of steam at this blasphemy. Hey, cut me some slack. If it wasn’t for the Hammond transplants, these Fishers would have been destined for scrap metal recycling. I’ve mentioned this before and I’ll mention it again, Japanese boutique amp builder Shindo uses Hammond OPTs in their least expensive (~$7,000 US!) amplifier. So there.
The reimagining of these amplifiers resulted in a higher parts count, so things got a bit cramped in the chassis. Also (gasp, horror) I didn’t stick to the original Fisher circuit (it was crap, see below). Changing the OPTs was just the licence needed to implement what I consider to be a better circuit, the SPP (Simple Push-Pull) used in my 2020 build series here on WoS. Ace opted for pentode connection of the EL84 tubes, like the original Fisher circuit. Pentode operation is explained later in this report.
The reimagining process, in brief, went like this:
-Removed the Death Capacitor (Death Cap (robrobinette.com) for an explanation).
-Removed all resistors and capacitors from the chassis and discarded them.
-Cleaned tube sockets and removed all bits of wire and solder from under-chassis sockets and lugs.
-Removed octal sockets used to power external preamps (Ace doesn’t have these).
-Ordered replacement resistors and capacitors.
-Discovered original OPTs are defective. (I should have known better and checked the trannies first.) I’d assumed Ace had run these amps, I guess not. I shouldn’t have assumed.
-Removed defective OPTs.
-Ace OK’d purchase of new OPTs.
-Waited for new OPTs to arrive.
-Removed previously cleaned tube sockets and ordered new ones.
-Made Delrin AC-in blocks that attaches in octal socket hole, see photo above.
-Made Delrin blocks to support new PS filter caps.
-Drilled mounting holes for new OPTs.
-Temporarily removed power transformers.
-Drilled holes in chassis to mount terminal boards and strips.
-Attached transformers, AC-in blocks, terminal boards and strips to chassis.
–IEC AC-in connector has ground pin connected to the chassis for electrical safety.
-Attached power switches, fuse holders and tube sockets to chassis.
-Installed small filter chokes for increased B+ filtering.
-Assembled new capacitors and resistors to chassis.
-Wired up the above including a LED to indicate power-on.
-Powered-up the first new “Fisher”, no smoke!
-My guesses as to new values for B+ dropping resistors were spot-on.
-Amp works but negative feedback connection was wonky.
-Corrected minor wiring error, feedback works as expected.
-Repeated the process on second amplifier.
And they sound like? Well, better than Fishers with defective transformers, that’s for sure. Amplifiers were driven directly by my PS Audio DirectStream DAC. The volume-adjust function on the PS was engaged. On the output side, my Triangle BR02s on homemade stands were connected to the amps with Kimber Kable. The Fishers had been burned-in for about 50 hours previous to the listening evaluation.
For a baseline I employed my prototype SPP EL84 amp (see What Difference Does An Output Transformer Make?) It is the second-best EL84 amp I’ve heard. The SPP was bettered, in small but meaningful ways, by the Finalé Audio F-7189 II reviewed back in 2019. To level the playing field as much as possible, I used the same EL84 and 12AT7 tubes in the Fishers and my SPP. I don’t mean the same brand; I used the very same tubes. I ran them in the Fishers, then pulled and inserted them into the SPP. I played a bunch of music: Some Keb’ Mo’, Loreena McKennitt, Beatles Abbey Road MQA stream, 24/192 classical ripped from LP, Steve Winwood, Bruce Hornsby and others as well. I won’t get into a track-by-track analysis as this article will be long enough as it is.
Sound Quality Evaluation, Take 1. Lovers of vintage tube sound might just forgive me for the liberties I’ve taken with these Fishers. Where my SPP sounds more immediate and attention-grabbing, the Fishers have a rounded and more forgiving presentation. The Fishers are smoother, especially in the mid frequencies. They might even satisfy my editor, he the lover of mellifluous midrange. The SPP is more detailed and extended-sounding in the high frequencies, but with “hot” recordings this can be a liability. With decent material the SPP’s HF detail can yield more specific imaging. The SPP has noticeably better pace and somewhat better bass extension. The Fishers, on the other hand, are more easy-going. A sizeable number of listeners would, I imagine, find this relaxing ease and flow more to their liking. My preference is for the snappier, more driven presentation of my prototype SPP. But I can appreciate some might find it a bit too intense.
Why the difference even though the reimagined Fishers and the SPP share the same basic circuit? They are very similar. Both have choke filtered and tube-rectified B+ supplies, though they use different rectifier tubes. Virtually identical Hammond output transformers adorn all amplifiers. The Fishers run their output tubes in pentode mode where the SPP is configured in Ultralinear, see description later in this article. The biggest difference I can think of is coupling capacitors. The SPP uses polypropylene film caps known for their speed and detail, whereas I had installed NOS paper-in-oil (PIO) capacitors in the Fishers. I can’t help but think that the Fishers lean a bit too far to the soft side and as a result are lacking in detail across the audio spectrum. To counter this but not to push them too far towards the sound of my SPP, I removed the PIO caps in the Fishers and purchased polypropylene film capacitors reputed to have an “analogue-like” sonic signature. (Editor’s note: I’ve found rectifiers also make a difference in how smooth/laid back an amp sounds.)
Sound Quality Evaluation, Take 2. The film caps were installed and the amps run for about ten hours to help them settle in. I configured the evaluation setup exactly the same as for Take 1. I ran through all the tracks mentioned above plus one more. Blue in Green, from Kind Of Blue, has some quiet and very delicate percussion brush work in the right channel. On systems with sufficient resolving power, it sounds like brush work. Systems, with what I would describe as having lower resolution, make the brush work sound like analogue tape hiss.
Before I go any farther, I’d like to define “vintage sound”. When I hear that expression, my first thoughts are of a warm but not especially detailed midrange. Then it conjures up full but not particularly detailed or fast bass and less than incisive treble frequencies.
From the above statement readers might gather I want something more than “vintage sound” from tube equipment in general and the Fisher 20-A in particular. If you disagree, you might want to stop reading now.
When I played Blue In Green, the Fishers, now adorned with modern polypropylene coupling capacitors, bested my SPP amp in subtle but significant ways. The brush work mentioned above was sufficiently clear with both the Fishers and the SPP. It wasn’t as distinct as I hear from my main system, but decent all the same. What I wasn’t expecting was the Fishers more burnished reproduction of Davis’s horn. On other recordings this subtle highlighting of midrange frequencies manifested itself in engaging ways. More expressive female voices, richer piano sound and slightly sweeter, more engaging, violin reproduction, to name the most obvious. None of these audible improvements in the midrange were at the expense of a reduction in high frequency detail. The Fishers gave up a bit of bass extension and force to the SPP. These differences were not huge but noticeable none the less. As far as sound-staging and imaging is concerned, the Fishers are now the equal of the SPP. This is an improvement compared to the PIO cap-equipped version in take 1.
All in all, I’m calling the change from PIO to film capacitors a resounding success. The increase in treble detail plus the slightly sweeter, but not unnaturally so, midrange frequencies is a win in my book.
Don’t hate me too much, vintage equipment aficionados. I think I’ve stayed true to the original Fisher essence. These amps are approximately six decades old. With a little TLC they might just see their eightieth birthdays, perhaps ninety or even one hundred.
Summing up, I’m sure the Fisher 20As don’t sound like they did when they were new. I think they likely sound better. And as a big bonus, in these times of supply uncertainty, they’ll be easier on their tubes now. (See below.) They have what I consider to be modern tube sound with just enough sweetness to maybe, possibly, hopefully, satisfy lovers of vintage equipment.
Next time: I tackle the recalcitrant Leak, again.
For circuit geeks (like me). A brief summary of changes implemented. See also the schematics below.
-Power transformers from half a century, or more, ago were designed to run on 115 volts. In many parts of North America line voltage can approach, and sometimes exceed, 125 volts. I added two 0.05-ohm resistors to the filament supply to protect the tubes from over-voltage.
-Filament supply connected to a portion of B+ supply instead of directly to chassis or circuit ground. It makes life easier on the 12AT7 especially and helps control ground loops, which cause hum.
-A 10-watt resistor was added to the B+ supply to drop the operating voltages to reasonable levels.
-Five B+ filter nodes implemented versus three originally.
-Grid stopper resistors added.
-Original gain/phase splitter tube was a 12AX7 (I’m not a fan). Changed to 12AT7 used in the SPP circuit.
-EL84s run at lower voltage and current to bring their dissipation into the range of 70% to 75% of recommended maximum. The Fisher circuit ran them at over 100%!
Pentode versus Ultralinear. The Fisher, above, has pentode-connected output tubes. The screen grids of the output tubes are supplied with voltage directly from a node in the B+ power supply. The SPP, shown in ultralinear configuration below, (it can also be configured for pentode or triode) has the screen grids supplied from taps on the output transformer primary.
Generally speaking, pentode is slightly more powerful but has higher distortion whereas, ultralinear has slightly lower power but also lower distortion. Again, very generally, distortion with increasing power (short of clipping) is more of a straight line in pentode. Ultralinear is lower but ramps up more steeply when approaching clipping.
Perhaps this is why virtually all tube guitar amps are pentode-wired creatures. Guitar players probably like their distortion delivered in a linear fashion. At least that’s my guess.
If I came across original Fisher 20-As at a yard sale, about fifty or sixty bucks for a pair is about all I would pay. There are just too many uncertainties with gear this old. Even if all transformers were good there would need to be a not insignificant investment in time and money to bring them to playable and safe condition. If you possess the knowledge and patience to do it yourself it could be very rewarding. Pay someone to do what I’ve done to this pair? You might want to pass. A vintage-equipment dealer would, no doubt, want a lot more money for a pair of these amps. A good look inside by an experienced tech would be essential before laying down any hard-earned cash.
Send a Donation
If you've enjoyed this content, consider supporting Wall Of Sound with a donation. It only takes a moment and will be greatly appreciated.