Let me rephrase the question: WHY should I not attach the negative feedback to the 4 ohm output, and what are the consequences if I do?
As @OldHector stated before: “The designer chose the 8R tap for calculating the amount of voltage to NFB.”
If you connect the negative feedback to the 4 ohm output the voltages generated will be insufficient to drive the NFB as designed, i.e. you will have less of it. Should not cause any instability but it will increase the sensitivity (amp will play louder at a given volume setting) and increase distorion. Try it if you wish, and see if you like reduced NFB from the 4 Ohm tap better than the NFB from the intended 8 Ohm connection.
If you connect the negative feedback to the 4 ohm output the voltages generated will be insufficient to drive the NFB as designed, i.e. you will have less of it. Should not cause any instability but it will increase the sensitivity (amp will play louder at a given volume setting) and increase distorion. Try it if you wish, and see if you like reduced NFB from the 4 Ohm tap better than the NFB from the intended 8 Ohm connection.
Perfect! Exactly what I needed to know. This amplifier build is supposed to serve as a learning tool for me, which is why I ask all of these questions. Thanks again!
Better safe than sorry, and it may not make a big difference, but only took a few minutes last night, so I matched all of the resistors and capacitors into precise pairs.
Unfortunately, if I use the PCB, that won't be possible. Testing and matching tubes using the technique below should help with the problem, correct?
This design is dependent on reasonably matched output tubes. Using the method suggested by Sorento you can switch your 4 tubes around for the best matching.
Maybe down the line you could try separate cathode resistors and bypass caps as an optimisation, if it is not a lot of surgery on the PCB.
Maybe down the line you could try separate cathode resistors and bypass caps as an optimisation, if it is not a lot of surgery on the PCB.
I most definitely will. EL84's are sold in matched pairs, so that might help in the future.
It is, unless I wire point-to-point and toss out the PCB. This amplifier chassis is small and very, very tight inside. Not much room to work with at all. The small size is one of its appeals to me.
Maybe I should take that back. I did a little math, and I found out that I was getting about 3,400 hours out of a quad of decent EL34 output tubes, which I have read isn't too bad? The Dynaco-branded tubes were awful, and they didn't last half that long. They also didn't have any bass, so the set that came in the amp and a replacement set they sent free when the amp blew up (burned resistors) were the last of those Dynaco-brand tubes to go in it.
Yes, I know this is off-topic for this thread. Sorry. As soon as I finish this build, I have learned so much from this thread, and one other thread, that I do intend to go through the Dynaco after 30 years of use to do some tests on it.
I bought a few little doo-dads for this kit, some not necessary but will make me feel better. I am going to change the 150Ω bias resistors to 160Ω per multiple suggestions. I also got cheap gold plated input jacks to help prevent corrosion. I am replacing the horribly old Philips 22uF 400V cap that is suspect due to age and the brand name. Two 100Ω resitors were recommended in the instructions but not included. Four resistors were playing a game of "which one of these is not like the others" with their mates in the kit and will be replaced. It would have been cheaper except for the $12 of shipping charges that I had to pay to ship $23 worth of parts - all total about $35 to make me feel better. I have all the wire I need on hand to upgrade it all to a slightly higher standard, so no cost there. Whenever the parts all come in (some from China, sigh) I can get out my soldering iron and get started. I'm glad I painted the chassis black. it looks much better, even if I leave the faceplate silver.
Last edited:
It's a question of general interest to myself and others, many of whom may not be reading this thread now or in the future, and it includes my Dynaco ST-70 series ii, not just this amplifier. Earlier you criticized my discussion of the Dynaco and a DAC in this thread, so take your pick. Can't have it both ways.
I never did change my tack, not then or now.
You don't even need to know the OT's primary DC resistance values. Instead, you'd put a current meter in the low milliamp range between both plates. Plate currents are balanced if the meter reads zero milliamps. This works well with a single pair of output tubes.
Best regards!
Now that is fascinating, although I did measure the OT primary DC resistance today for the new amp kit as it only took a minute since the amp isn't assembled yet. I have no idea what the values would be on the Dynaco ST-70 series ii unless they are published online somewhere, but using your technique, I do not need to know.
I am considering drilling a few holes in the chassis of the new amp for test points where I can insert probes. Otherwise, it involves removing it from the system (a pain) and then removing a lot of screws to get to any of the test points. I assume Mouser or Digi-Key, (or ebay or AliExpress) sells a part appropriate to the task, but I don't know what it is. No experience there, but a lot of amplifiers have test points, so the parts must be available.
Last edited:
Why have they used 5 watt resistors for R21 and R22? Isn't that very much overkill? There are only supposed to be 10 volts and about 60mA there? Schematic is in the first post.
60mA is the quiescent current for two tubes. Over 150R that gives a voltage drop of about 10V.
Assuming the amp designer knew what they were doing (not a bad idea), and allowing for 3 times safety margin, then due to peaks from the signal voltage, the tube is turned on more, current increases, voltage drop increases, and dissipation (power consumed in the resistor) increases. Maybe 5W is marginal in the extreme case.
Adding to that, an oversized resistor will not get that hot, more area, better dissipation. Lower temperature inside the chassis is always better.
As the total warmth energy remains the same, resistor size doesn't affect the temperature beneath the chassis.
Best regards!
Best regards!
Ah yes, that explains it. Thanks!
Based on the recommendation of three people, I am changing those resistors from 150Ω to 160Ω. If something doesn't sound right, I can always put in the 150Ω as designed. They are easy to change.