 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
76.127.54.176
In Reply to: RE: Building clone amp - question about grid resistor and coupling cap posted by PeterI on June 21, 2021 at 09:25:55
then DON'T change the value of the grid to ground resister very much as that will not change the frequency passed to the output tubes as much.
That said, the existing values calculate out to pas about 20Hz (see link below) which is likely lower than the output transformer would be able to do too much with.
Are you using original iron?
The HF81 used GREAT iron, and if I were to do what you are doing I'd look for a trashed HF-81 with original, and still working, iron as that was the secret to the HF-81's great sound.
That and the EL-84 being a great sounding tube.
HF-81 iron will also work using the 6V6 tube as well.
Follow Ups:
Thanks for that info. Actually, I'm imitating the Eico HF81 output, same as the HF14 monoblock amp. I've owned, worked on and restored 4 HF81s (now sold). I have some pretty good vintage OPTs. I guess I'll stay close to original R and C values.
If I find another trashed HF81 (a hard to find item now), I'd probably restore or rebuild, as I have before. Fun to work on.
Cheers
P
Looks to have been run over by a truck, but the iron checks good.
Was thinking to built a power amp out of the iron (both power and output) by merely substituting octal tubes (while adding a 6 V filament transformer) and following the HF-86 EICO shematic.
Circuit would be ideal for 6SL7 subbing for 12ax7 section and 6SN7 for 12AU7 section of the 7247.
Lot of octal rectifiers to choose from.
And the 6V6 is maybe the best sounding of the whole octal power tube lot, in the right circuit.
And I agree, the 6V6 makes great sound.
Cheers
-Pete
Babylonart@aol.com
If these amplifiers are biased AB1, none should be rebuilt with self bias. As I wrote recently in another thread, DC voltage across the self-bias network increases at higher amplifier outputs and anode currents. That causes bias voltage to increase, moving the output stage closer to Class B. Unless the bias is set very high at idle, this movement creates crossover distortion that is easily observed on a scope. This distortion will persist into quiet musical passages due to the time constant of the bias circuit, muddying nuance and degrading fidelity.
The only reason manufacturers ever got away with this is that printed specifications minimize the impact. At higher amplitudes, the crossover distortion represents a numerically small percentage that simply adds to the naturally increasing distortion levels. In addition, the loudness of the system at higher levels tends to hide what's happening from the listener.
These older designs can be greatly improved with either the "Enhanced Fixed Bias" cathode bias technique described at the tronola.com site, or with the addition of fixed grid bias. This should not be neglected if you want your restoration work to be fully worthwhile.
--------------------------
Buy Chinese. Bury freedom.
As a bit of an aside on the EFB...
Inmate DAK reported a failure of the LM337 when used in the EFB circuit, in a post located Here .
In a subsequent post, I commented that I use EFB in an SCA-35 and hadn't had any issues with the LM337, which is still the case.
I'm starting to re-explore ideas for a long mothballed ST-35 clone, and I'm planning to use EFB in that too. I happened across a post from Dave Gillespie (link below) where the "handful" of LM337 failures have been addressed.
If anyone is considering (or already using) the EFB, it appears to be valuable information regarding protection of the circuit.
I wasn't aware there had been failures, and I haven't analyzed the circuit except in a general conceptual context. Three things to note...
First, Dave hasn't been able to duplicate the failures. That's a big problem, because it leads to revisions that may or may not be relevant or effective. For all he knows (and I'm sure he would agree), it's possible every one of the failures was due to the use of counterfeit parts from China that were purchased through disreputable sources such as eBay or Amazon.
Second, recommendations in the datasheet regarding protection diodes and decoupling capacitors should be all that's required to fully bulletproof the IC. I've designed quite a few Vregs and applied three-terminal regulators in a wide variety of applications. When it comes to these decades-old devices, the manufacturers pretty much have it nailed. Although I frequently refer people to Dave's site, I would never use the LM337 myself without following those guidelines.
Finally, the filaments should be elevated using their own voltage divider from B+, regardless of the method used for output tube bias. This not only eliminates the issue Dave theorizes regarding damage to the regulator, but allows raising the differential to a level more likely to provide the desired benefit.
--------------------------
Buy Chinese. Bury freedom.
When I posted that link to AK, I had yet to thoroughly read the thread, but I did so last night. Gillespie states that a traditional diode between the OUT and ADJ terminals of the LM337 will protect against "stress mode #2".
But he further states, "Resolving the 1st stress mode however requires the diodes to be installed rather differently, because here the problem is not one of reverse polarity, but of an over-voltage condition. The normal operating voltage between the ADJ and OUT terminals is ~ 1.20-1.25 vdc, with the exact value being dependent on the manufacturing tolerances of the particular regulator piece used. If two small signal diodes were connected in series and then strapped across the ADJ and OUT terminals -- opposite in polarity to the reverse polarity diode, then some regulators might work, but many might not because the voltage drop across the two diodes is virtually the same as the normal voltage drop produced by the regulator itself. The use of just two diodes in series then could interfere with normal regulator operation.
To guard against that from happening, a string of three diodes is used, all connected in series, connected across the two regulator terminals as described. Now the diodes prevent the over-voltage condition from reaching beyond about 1.80 vdc (total), but are completely disengaged during normal operation, since the normal drop of the three diodes well exceeds the normal operating voltage produced by any example of these regulators between these two terminals."
For the cost of four diodes, I'm not going to quibble with his test results.
Well, Dave is correct that the IC requires a certain minimum current flow in order to provide regulation at the output. However, I'm not sure I agree that leaving the output open (as happens prior to tube warmup) is a stressor. There are no warnings about this in any 337 datasheet that I've ever seen, nor are there any reports of failure from other segments in the field AFAIK regarding any permanent damage that might result.
That aside, it does no harm. Three series diodes connected with the anode on the ADJ pin and the cathode on the OUT pin will ensure that ADJ can never be more than 1.8V positive relative to the output.
Incidentally, I'm not surprised his buddy couldn't duplicate the problem in simulations. I remember discovering a few years ago that some opamps have output with no VCC applied. Not very helpful when you're troubleshooting!
--------------------------
Buy Chinese. Bury freedom.
Given that it'll do no harm, and it's inexpensive, I'm going to add it to my circuit. Here's the revised EFB schematic with the diode orientation, courtesy of Dave Gillespie.
If you use an LED in place of the three diodes, it should flash when that condition exists, right? :^) I think orange or amber would be about 1.8 volts.
That would be cool! The positive voltage to the ADJ pin is derived from a 360K ohm in series with B+, so there's less than 1 mA available. Kingbright makes LEDs rated for 2 mA that might be visible. Vf at 2 mA is about 2V.
--------------------------
Buy Chinese. Bury freedom.
I'm going to have to do some reading about the how and why of elevating filaments.
Below is a link to a more extensive discussion of DAK's LM337 failure in the EFB circuit. You suggested a protection diode per the LM337s data sheet, but I have yet to heed it. My SCA doesn't see much use, perhaps only a dozen hours since this discussion in 2018, but I've never had a bit of trouble with the LM337 in all the years its been in there.
I'm currently rebuilding the power supply of the SCA-35, as well as populating and installing Dave Gillespie's replacement amp boards, which optimize the circuit to use 6GH8A on the input instead of the previous 7199.
Since I'm in there anyway, now would be the time for me to add the diode to the EFB.
"Below is a link to a more extensive discussion of DAK's LM337 failure in the EFB circuit. "
LOL, and I was one of the minor contributors. I had forgotten all about it.
FWIW, there are other techniques for fixed cathode bias. I designed a simple circuit some time back using a darlington transistor and a handful of resistors. It's not as "perfect" as the LM337, and it doesn't account for changes in B+, but it's guaranteed to be bulletproof. Actually, a direct anode-cathode short might do it in, but things like that are always a risk when we mix SS with tubes. Performance wise, a simple technique like this is sufficient to hold the cathode within about 5% of the set voltage, which is a big improvement from the 50% or so it can move otherwise.
I agree, it's a good idea to add the protection diode(s), even if the regulator appears to be working well.
--------------------------
Buy Chinese. Bury freedom.
I had a vague recollection of our exchange, but couldn't find it when I did my initial search. I'm often surprised when I read old posts I contributed to, but have no recollection of. The good news (for me, anyway) is that I still often agree with myself. Or maybe that's not so good...
Thanks for that advice. I always look at Dave G's designs though much content in his articles are over my head. I have used fixed bias in other builds, using this design borrowed from Jef Larson for his 6V6 amp.
The EFB is partly designed to stabilize a stereo amp with 2 channels competing for power. My project is for mono amps with LOTS of current headroom. Will certainly try this if I have the real estate to fit a small board.
Thx
-Pete
That's similar to the bias circuit I designed for my own amplifier project, currently in process. Use of FW rectification proved advantageous regarding ripple during my simulations. Some people say this isn't overly important, because the push-pull output transformer doesn't respond to common mode energy. However, I've always believed the 60 Hz ripple amplified in-phase by each output tube can indeed become audible as IM distortion on peaks. Like so many other things, this doesn't appear in spec sheets. :)
--------------------------
Buy Chinese. Bury freedom.
I see he supplies the EFB from the B+ of the SCA-35. ~350VDC? into a 2w 360Kohm R.
I'm not sure I'd know how to adapt that to my HF-14 clone. He doesn't give voltages to the in and out of the LM337.
The datasheet for the LM337 has the formula for calculating output voltage. His values will probably be very close though. Tweaking the 15K should be enough to put the adjustment pot near the middle.
--------------------------
Buy Chinese. Bury freedom.
nt
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]