The cathode current of every valve (as measred with a current meter in series), even a low signal level class A triode, varies somewhat with signal level, because they're not perfectly linear. For push pull output valves the region where both valves contribute to output current has a comparatively small increase in current with increasing signal (but still some) and signal regions into the class AB operation have larger current variations.
Cathode resistor bias usually (not always; the original DTN Williamson did not - very elegant but class A only) has a large capacitor bypassing the cathode resistor. So, bias voltage changes dynamically with signal level, with a time constant RC where R is the cathode resistor in parallel with reciprocal of transconductance of the valve (the resistance "looking into the cathode") and C is the bypass cap. An active current source like D. Gillespie is effectively a very large resistance and R of the time constant in that case is only the reciprocal of transconductance.
Various arguments have been made over the years favoring common cathode resistors - Dynaco in particular made a decent case - all center around the smaller variation in bias when one channel is driven and the other is not. None consider time constant or OPT distortion from unbalanced primary currents. Ya pays yer nickle and ya takes yer choice.
If you worry too much about this issue you end up using adjustable "fixed" bias instead. It's easy enough, but is it worth it?
All good fortune,
Chris
Cathode resistor bias usually (not always; the original DTN Williamson did not - very elegant but class A only) has a large capacitor bypassing the cathode resistor. So, bias voltage changes dynamically with signal level, with a time constant RC where R is the cathode resistor in parallel with reciprocal of transconductance of the valve (the resistance "looking into the cathode") and C is the bypass cap. An active current source like D. Gillespie is effectively a very large resistance and R of the time constant in that case is only the reciprocal of transconductance.
Various arguments have been made over the years favoring common cathode resistors - Dynaco in particular made a decent case - all center around the smaller variation in bias when one channel is driven and the other is not. None consider time constant or OPT distortion from unbalanced primary currents. Ya pays yer nickle and ya takes yer choice.
If you worry too much about this issue you end up using adjustable "fixed" bias instead. It's easy enough, but is it worth it?
All good fortune,
Chris
Nice work on the extraction- any Dentist would be impressed... Either you have very big claws, or some type of mechanism.
I cheated by cleaning out my cans w/ a Bridgeport. Ordered Rubycon TWX skinnies to fill. Now wondering what to use as filler- and how to provide terminals. Current idea is RTV silicone to fill the body. Once cured, 2 part hi temp epoxy surrounding each section (+) male spade lugs. (-) combined as single point lug, since can't solder to AL
Jim
I solved the problem of good, solid terminals by retaining all of the originals, including the twist tabs that hold the entire cap in place. I did this by slowly grinding away the crimped edge of the can with a Dremel tool sanding drum attachment instead of trying to pry it open or cut it open. Once the crimped edge is ground away, the whole guts including the terminals and base can be pulled out of the can in one piece. I did this by putting in the oven to melt the tar. I think it was 20 minutes at 200 degrees, maybe 225. Then I unrolled the cap and trimmed off the internal electrodes flush with the bottom piece. This left the bottom piece 100% undamaged. Slowly grinding away the crimp instead of prying it open left a very clean, perfect edge on the can and no damage to the terminals or twist tabs so the bottom piece is still in new condition and ready to use.
There was a lot of tar inside the can, but it's easily removed I discovered. Simply fill the can with mineral spirits and let it sit for 12 hours. Dump it out, wipe with paper towels, and repeat once or twice until all of the tar is gone. Mine came totally clean.
I don't plan to fill the inside of the cap can. I want it to be removable so I can change caps if I ever need to. You never know when disaster might strike. I'll use the twist tabs to attach the cap base and terminals, with the new caps attached to that. Then I'll slip the original can over it and secure it to the base with something that is relatively easy to break loose like a thin film of RTV silicone.
Good point about not filling-
My can had bulged out the bottom, ESR tested bad all sections. Also- didn't have bottom bakelite- circa 1948... Still admire your tenacity.
Will add that I found good source for new 4 section caps here:
https://www.ebay.com/itm/130326876069
Ordered 1 for diff project- new- tested strong all sections.
My can had bulged out the bottom, ESR tested bad all sections. Also- didn't have bottom bakelite- circa 1948... Still admire your tenacity.
Will add that I found good source for new 4 section caps here:
https://www.ebay.com/itm/130326876069
Ordered 1 for diff project- new- tested strong all sections.
To avoid any problems, I bought 2 new old stock quad caps still in boxes on Fee-Pay for about $12 each, shipping included, so there would be no damage to them. They had stickers on the cans showing the values instead of stamped lettering on the can. I wanted stickers that I could peel off since the values will no longer be correct. I'll be painting the can black with a white decal showing the new values and the date. I am working on new decals for the whole amplifier now, since I have started painting the chassis black. Since the amplifier has a number of changes compared to the original, including 8/4 ohm transformers instead of 16/8, feedback moved to the 8 ohm tap, changes to the power supply, and the add-on adjustable bias board, I'll just call it the Stereo 35 Series ii similar to my Dynaco (Panor) Stereo 70 Series ii. When Dynaco put "power amplifier" on their equipment, they used ALL CAPS. I think I'll modernize to upper and lower case.
I'm still using the same set of tubes and the bias has held steady. For the first year or so I ran it without the cage like I do my VTA ST-70. But I did go and buy the cage awhile back. With the cage on, the sides of the cage do get pretty hot since the sides aren't perforated. But to me the amp is just too ugly to look at without it.
At first I bought some 3/4" rubber feet for it, then I saw a picture of these amps posted on another forum with the spikes. Personally I thought the spikes looked good on it, so I bought a set from Madisound. You have to use the flatter pan head screws though so nothing gets shorted out inside the amp.
In "the old days" a large portion of Push Pull vacuum tube kits came with . . . Guess What?
Very well matched output tubes.
That allowed the manufacturer to do two things to save cost of some other parts in the kit.
1. A single cathode self bias resistor & single bypass capacitor that was connected to all 4 cathodes (push pull stereo).
2. A slightly smaller output transformer with less laminations, because the push and pull output tubes plate currents were very well matched,
(there was no significant difference in the quiescent DC current in the push and pull primary halves).
Bonus: The single self bias resistor and single bypass cap also made for simpler assembly and a smaller sized amplifier.
. . . Practical Engineering that made for very good kits versus the cost.
There I said it.
Very well matched output tubes.
That allowed the manufacturer to do two things to save cost of some other parts in the kit.
1. A single cathode self bias resistor & single bypass capacitor that was connected to all 4 cathodes (push pull stereo).
2. A slightly smaller output transformer with less laminations, because the push and pull output tubes plate currents were very well matched,
(there was no significant difference in the quiescent DC current in the push and pull primary halves).
Bonus: The single self bias resistor and single bypass cap also made for simpler assembly and a smaller sized amplifier.
. . . Practical Engineering that made for very good kits versus the cost.
There I said it.
True, but even with starting with matched pairs, or quads, over time they may not age evenly. I always start with matched tubes, whether it's a single ended amp, PP, or PPP, no mater the bias scheme.
Ditto. Also, different sellers, especially overseas sellers of the 6P14P-EV, have different matching standards, and I don't always trust their accuracy.
I'd like to open up the sides of the cage instead of leaving it solid sheet metal. Does anyone have, or know where I can get, a small amount of the perforated metal cage material? It would be very easy to cut out the sheet metal sides with a Dremel tool and use JB Weld to put mesh in their places.
Some people love the look of tube amps without the cage. I am one of those, but I also can't stand silver audio components. I bought a Chinese amp kit in January that was all silver, and it's now all black. This ST-35 will be all black as well. I painted the choke and cap can black yesterday, and I got the first coat of black on the chassis and the bottom also. They need wet sanding followed by a finish coat and a few weeks of drying time. The new white decals are finished and almost ready to send to the printer.
I'll have to find some better feet. I'd prefer rubber vs. damaging spikes. I don't mind that the amp sits so low, and with it being open, I can't imagine that ventilation is a big issue, but I want to be able to get my fingers under it when I need to move it.
I'd like to add a power switch on the chassis instead of the inline version included. I'll have to find a suitable small switch. It seems like the least bad option for placement is on the back as the person in the picture below did. I could put it on the front I guess. I wish there was a tiny rocker switch that could be used instead of a toggle.
I'm not sure why this person chose to use a 2-wire power cord instead of a modern 3-wire cord with safety ground. I suppose I'll use 3-wire with a star grounding pattern. I'm told that the isolated input jacks have to be connected to a ground.
I'll probably fabricate speaker binding posts properly spaced for dual banana plugs. I'll just connect the 8 ohm output and leave the 4 ohm output wire coiled inside in case anyone ever wants it, but I highly doubt it. On both of my amps, the 4 ohm output sounds terrible.
An inrush current limiter can be be incorporated to soften the power on process and thus reduce the potential damage of sudden full voltage being applied to the cold vacuum tubes.
Thoughts on this anyone? Would this (CL-90) be the correct item as this ST-35 owner did in the photo? https://www.mouser.com/ProductDetail/Amphenol-Advanced-Sensors/CL-90?qs=K5ta8V%2bWhtYTYrFAkIfaLQ==
I have to wonder if putting this device inline with the supply of a power amplifier would cause more harm than good? A lot of people seem to do it though.
Thoughts on this anyone? Would this (CL-90) be the correct item as this ST-35 owner did in the photo? https://www.mouser.com/ProductDetail/Amphenol-Advanced-Sensors/CL-90?qs=K5ta8V%2bWhtYTYrFAkIfaLQ==
I have to wonder if putting this device inline with the supply of a power amplifier would cause more harm than good? A lot of people seem to do it though.
Last edited:
I just by rubber feet from Amazon. I've bought 1/4", 1/2", and 3/4" tall feet there.
CL-90's are what I usually use. As for the switch, there isn't much room in that chassis. I bought a Bluetooth remote controlled outlet and use it for my ST-35. It works great. I'm not a fan if the wifi power strips though.
I had one flip out and turn two stereo tube amps and a pair of mono blocks on at the same time when I selected one if the amps. It fried the power strip. I've had zero issues using Bluetooth outlets on various amps. I don't like having to reach behind hot amps to turn them off.
Last edited:
Inrush current limiter. Completed - well less faceplate - a ST35 build a few weeks ago. Dynakitparts transformer set in one of their SCA35 chassis for extra space. Dave G’s boards including the EFB one - I’m pleased with it but to each your own of course. Used a CL-90 in it although I believe a CL-80 would be fine also. Just seemed like a sound idea, had the room for a 2 position barrier terminal strip and had seen them in a number of Pass clone builds.
CL-90 ... not sure if this does any good for the tubes since the filaments are going to take 15-30 seconds to heat up, long after the inrush. Seems like this would only be of value to (perhaps) protect large power supply capacitors. Not sure how this works.
That tubes ages differently is a myth. In most cases they will degrade ( aka age) in same pace. That is
that they are same brand and same age of course.
I will be putting that to the test as this amplifier allows measuring and adjusting all four tubes independently. If I never have to adjust anything, great, but the option is there, and I will record the actual results over time.
This happened to me so it's not a myth - but my case is a little different - tubes run on maximum dissipation - the box is a closed type with a lot of perforations which are not enough and everything is done on big motherboard.
So take into account overall conditions - tubes are cooking themselves and everything else besides.
To be honest different ageing happened only on one pair ( one channel ) out of three quartets used overall.
I guess it's not common but can happen.
My point was completely about the old kits and matched tubes:
The Old kit manufacturers used tubes of the same brand and age that had been carefully matched, probably most usually that were matched in a built amplifier of the same version as the kit right on the kit manufacturers premisise.
We are talking about the 1950s and 1960s.
Things were different back then.
The matched tubes I use today are JJ tubes from Eurotubes.com
Eurotubes tests lots of different characteristics. They are re-tested on 2 or more test fixtures.
Works for me. They age well together.
But, I do not abuse those tubes.
Some Hi Fi amplifiers and some guitar amplifiers abuse the tubes, in that case Your (Their) Mileage May Vary.
Mistreat your pet goldfish, cat, or dog.
You will have to replace it.
The Old kit manufacturers used tubes of the same brand and age that had been carefully matched, probably most usually that were matched in a built amplifier of the same version as the kit right on the kit manufacturers premisise.
We are talking about the 1950s and 1960s.
Things were different back then.
The matched tubes I use today are JJ tubes from Eurotubes.com
Eurotubes tests lots of different characteristics. They are re-tested on 2 or more test fixtures.
Works for me. They age well together.
But, I do not abuse those tubes.
Some Hi Fi amplifiers and some guitar amplifiers abuse the tubes, in that case Your (Their) Mileage May Vary.
Mistreat your pet goldfish, cat, or dog.
You will have to replace it.