"It is only 12AT7 and not 12AU7, pay more attention next time you read someones schematics."
My apology, the small print on the simulations posted was hard to read. I think where I picked up ECC82 (12AU7) was here for the RH88:
http://www.diyaudio.com/forums/show...392#post1739392
"Have you ever tried a real FFT analyser?"
Yes, I have Tek and HP analysers. They all agree remarkably well with the PC analyzer. I did upgrade to a 24 bit M-Audio sound card when I realized how powerful FFT analysis was. It's a Delta 1010 unit with the D/A's and A/D's in a separate box to stay away from PC digital noise.
"You read correctly the part about the punch, but the solid bass is not a consequence of slightly higher 3rd harmonics or some mumbo jumbo, but plain damping factor, increased by adopting this form of feedback."
Yes, damping factor should be improved by shunt feedback as you say. But a little 3rd H of the correct phase (expansive) helps too.
"we are talking here about SE pentode"
Yes, I should have distinguished. I do recall your commenting in the other thread that RH84 was intended to compete with 300B SET designs but at lower cost. I also recall a comment to the effect that the large screen grid resistor value enabled the screen grid to track the plate voltage.
Don
My apology, the small print on the simulations posted was hard to read. I think where I picked up ECC82 (12AU7) was here for the RH88:
http://www.diyaudio.com/forums/show...392#post1739392
"Have you ever tried a real FFT analyser?"
Yes, I have Tek and HP analysers. They all agree remarkably well with the PC analyzer. I did upgrade to a 24 bit M-Audio sound card when I realized how powerful FFT analysis was. It's a Delta 1010 unit with the D/A's and A/D's in a separate box to stay away from PC digital noise.
"You read correctly the part about the punch, but the solid bass is not a consequence of slightly higher 3rd harmonics or some mumbo jumbo, but plain damping factor, increased by adopting this form of feedback."
Yes, damping factor should be improved by shunt feedback as you say. But a little 3rd H of the correct phase (expansive) helps too.
"we are talking here about SE pentode"
Yes, I should have distinguished. I do recall your commenting in the other thread that RH84 was intended to compete with 300B SET designs but at lower cost. I also recall a comment to the effect that the large screen grid resistor value enabled the screen grid to track the plate voltage.
Don
Re: CHEAP FFT ANALYSERS
I thought you knew that the cheapest massproduced soundcard today have a pretty descent 24bit A/D converter. To use a soundcard as a scope is most often not a choise but as a FFT analyzer ... yes, very good with the right software. I guess you know that most scopes have a resolution of 10 to 12 bits?
[/B] [/QUOTE]
The FFT thing is the best that happend DIY's with a PC and a descent soundcard. Not scope or other things which demands high samplerates but not high resolutions. 192khz 24bit as an FFT hardware is hard to beat. If you build a frontend amp to separate the measuring and protect the card, you have something really good to work with.
I can tell you that I work as a metrology engineer at an Instrument calibration maintenance dep. on a chemical industry. I have not seen it all and I'm not used to older FFT equipment but I have seen a lot of expensive instruments. I definately doesn't "diss" a descent "cheap" PC card as an FFT analyzer.
Alex Kitic said:
I thought you knew that the cheapest massproduced soundcard today have a pretty descent 24bit A/D converter. To use a soundcard as a scope is most often not a choise but as a FFT analyzer ... yes, very good with the right software. I guess you know that most scopes have a resolution of 10 to 12 bits?
[/B] [/QUOTE]
Again, cheap can be good. Tell us about your experience on "real FFT's" How old was that instrument and what was the resolution on the A/D converter.
The FFT thing is the best that happend DIY's with a PC and a descent soundcard. Not scope or other things which demands high samplerates but not high resolutions. 192khz 24bit as an FFT hardware is hard to beat. If you build a frontend amp to separate the measuring and protect the card, you have something really good to work with.
I can tell you that I work as a metrology engineer at an Instrument calibration maintenance dep. on a chemical industry. I have not seen it all and I'm not used to older FFT equipment but I have seen a lot of expensive instruments. I definately doesn't "diss" a descent "cheap" PC card as an FFT analyzer.
A new thread??
Hello folks:
I have been reading this thread and was quite enjoying the dialog and experimenting with the design brought forth by Ian.
I am quite concerned that this current direction has lost the intent and moved well away from the original thread.
Perhaps you could take your discussion about the RH concept and start another thread so the discourse can return to the discussion about the Baby Huey.
Hello folks:
I have been reading this thread and was quite enjoying the dialog and experimenting with the design brought forth by Ian.
I am quite concerned that this current direction has lost the intent and moved well away from the original thread.
Perhaps you could take your discussion about the RH concept and start another thread so the discourse can return to the discussion about the Baby Huey.
I am sorry the RH84 sneaked in, but is has some relevance as the discussion is about plate to plate NFB. Think it started when Alex dissed my suggestion of another kind of plate NFB of the Huey.
Sorry again, shouldn´t have replied to the accusations. that obviously poured more gasoline on the fire
!
So back to the original subject with PP with local NFB! I am sure the Huey can get even better with tweaking of the NFB.
Sorry again, shouldn´t have replied to the accusations. that obviously poured more gasoline on the fire
! So back to the original subject with PP with local NFB! I am sure the Huey can get even better with tweaking of the NFB.
PATIENCE
Dear all,
Thanks for your patience. I excuse myself for stealing your thread for a period of time, but it seems that I am not the only one to blame. I just did not want to let other finish it, and that I admit being guilti for.
Point is, the plate to plate feedback is a great way to improve the sound and performance of an amp, and is fully applicable to PP amps. The old schematics shows you that it was done a long time ago... but the 2meg resistors mean that the concept was not realised in the same way we (I) do it nowadays.
I have no direct experience with FFT analysers, as I did never have the need to. In the end it turned out that PC FFT is the best thing that happened to those who need FFT, and everyone eventually boasted out their expensive sound cards... I imagine the author of the TL082 based buffer amp (now some could start telling us that the TL082 is a no-no in audiophile terms, just like the "mixer ready" ECC81, could they) back in 1996 had a state of the art sound card with 24bit 192kHz resolution? If you have enough money to buy a 1010, you could buy the "real thing" as well, since you have already crossed the point of diminishing returns.
You see, I am not a technical person by profession, actually I am an economist and do some trading, logistics... etc. for a living. That is why I take costs and results into account.
Excuse me one more time, and feel free to invite me to discuss your new ideas, or my previous designs.
Regards to all,
Alex
Dear all,
Thanks for your patience. I excuse myself for stealing your thread for a period of time, but it seems that I am not the only one to blame. I just did not want to let other finish it, and that I admit being guilti for.
Point is, the plate to plate feedback is a great way to improve the sound and performance of an amp, and is fully applicable to PP amps. The old schematics shows you that it was done a long time ago... but the 2meg resistors mean that the concept was not realised in the same way we (I) do it nowadays.
I have no direct experience with FFT analysers, as I did never have the need to. In the end it turned out that PC FFT is the best thing that happened to those who need FFT, and everyone eventually boasted out their expensive sound cards... I imagine the author of the TL082 based buffer amp (now some could start telling us that the TL082 is a no-no in audiophile terms, just like the "mixer ready" ECC81, could they) back in 1996 had a state of the art sound card with 24bit 192kHz resolution? If you have enough money to buy a 1010, you could buy the "real thing" as well, since you have already crossed the point of diminishing returns.
You see, I am not a technical person by profession, actually I am an economist and do some trading, logistics... etc. for a living. That is why I take costs and results into account.
Excuse me one more time, and feel free to invite me to discuss your new ideas, or my previous designs.
Regards to all,
Alex
Baby Huey
There is probably a lot of people building this Baby Huey that eventually bring there questions or findings up here.
I have gathered a lot of the parts and will build it but my projects take decades so...
I have seen other threads with groupbuys that take a stop. I am, so far, the only one makin' the Kleinschmidt 10W class A amp for instance. There is a lit of PCB's to that amp thats not yet populated. People have a lot of hobbies but not the time they need.
I have a headphoneamp I'm bankin' on but this weekend I had to take care of a water leak and also buy a new dishing machine.
And then the old 4:3 TV set, "thick CRT TV", went dead, so we also bought a new 42" Panasonic plasma that I have to fiddle with a bit...
And then there is the spring with all apple threes that no one in the family had cut properly. And the car needs some washing now and then. Then we have the house ...
And our daughter need some attention with her horse and we also need to repair the horse transport and then there is my own parents and there need of help to move to a new appartment from the old house ...
Then the work to get the living everyday day.
The Baby Huey I think have alot of fans working on their version but it takes time.
This thread have had some pauses before this last discussions, so don't be sorry Alex.
There is probably a lot of people building this Baby Huey that eventually bring there questions or findings up here.
I have gathered a lot of the parts and will build it but my projects take decades so...
I have seen other threads with groupbuys that take a stop. I am, so far, the only one makin' the Kleinschmidt 10W class A amp for instance. There is a lit of PCB's to that amp thats not yet populated.
People have a lot of hobbies but not the time they need.I have a headphoneamp I'm bankin' on but this weekend I had to take care of a water leak and also buy a new dishing machine.
And then the old 4:3 TV set, "thick CRT TV", went dead, so we also bought a new 42" Panasonic plasma that I have to fiddle with a bit...
And then there is the spring with all apple threes that no one in the family had cut properly. And the car needs some washing now and then. Then we have the house ...
And our daughter need some attention with her horse and we also need to repair the horse transport and then there is my own parents and there need of help to move to a new appartment from the old house ...
Then the work to get the living everyday day.
The Baby Huey I think have alot of fans working on their version but it takes time.
This thread have had some pauses before this last discussions, so don't be sorry Alex.
Alex etc.
Yes - we want discussion and ideas. As soon as I finish the guitar amp I'm working on I will be back on the Baby Huey amps for another round of experiments. Meanwhile, the few weeks break just listening to them is worthwhile as it pins down exactly what aspects of the sound I'm really happy with and what aspects I'd be happy if I could improve. I have printed the last 6 or 8 pages of the thread and I will try a few of the ideas, in particular a couple of the things that Lars modeled - why because I think the amp may have a touch too much distortion BUT I don't want to add any more feedback which may possibly destroy the immaculate timing.
Cheers,
Ian
Yes - we want discussion and ideas. As soon as I finish the guitar amp I'm working on I will be back on the Baby Huey amps for another round of experiments. Meanwhile, the few weeks break just listening to them is worthwhile as it pins down exactly what aspects of the sound I'm really happy with and what aspects I'd be happy if I could improve. I have printed the last 6 or 8 pages of the thread and I will try a few of the ideas, in particular a couple of the things that Lars modeled - why because I think the amp may have a touch too much distortion BUT I don't want to add any more feedback which may possibly destroy the immaculate timing.
Cheers,
Ian
Re: LAST TIME, NO USE, OBVIOUSLY!
Yet there are tons of posts on these very forums saying that the total number does not matter if you do not know the harmonic distribution. It is a simple matter looking at the amount of odd harmonics in the simulator, with a dB scale - unless you are using truly high screen resolutions, harmonics 40-50dB below the fundamental (0.5-1% distortion) will not be seen on a linear scale, regardless of the order.
If you have used something like curve captor to get a realistic model, they will show up at higher power levels. Pretty much any SE amp that I have ever seen, or at least read the relevant documentation about, exhibits this distortion - RHxxx will not be an exception. After all, where do you think the odd harmonics come from in a PP amp? They are not a magical property of the PP principle, but appear because PP is not able to cancel them out unlike even harmonics, so thay had to be made in the stages of the amp in the first place - and pretty much every PP amp can be split into two SE equivalents.
Interesting that it is you calling what I write 'mumbo jumbo' - this is a first for me and I have never done so to you. But, you have asked a question:
If single stage of amplification produced only the second harmonic, two cascaded stages will produce harmonics up to 4th, by the process of harmonic multiplication - simple, really, the 4th harmonic is the second harmonic generated in the second stage, of the second harmonic generated in the first. Similairly, if a single stage generates harmonics up to 3rd, the output will contain 2nd, 3rd, 4th, 6th, 9th. Of course, now it is a question of how you define 'higher order', and of course what amounts we are talking about, but 9th would certainly qualify.
All of this is not accounting for various methods of applying feedback, or nonlinearities of other components, mostly the output transformer.
It shouldbe noted that two inverting cascaded stages may be set-up so that inverse distortion cancellation takes place. However, it is important to know that it will cancel most of the second harmonic, but may still generate the 4th harmonic by harmonic multiplication, which is quite counter-intuitive. In theory, if the stages were completely equal, all even harmonics could cancel out, but in cases where we are speaking about a driver and output tube, this is so difficult it may as well be impossible - it may happen with a particular tube, but will not be repeatable with other examples of same tube type and probably will not hold as tubes age. Setting non-matching operating points in the square law transfer characteristic may result in lower 2nd harmonic due to cancellation, but the non-aligned curves will produce a composite transfer characteristis with a 4th order nonlinearity.
Because the human ear automatically supresses harmonics that have a monotonically decaying (on a dB scale!) amplitude with rising order, canceling out only the 2nd harmonic but leaving 4th (or 3rd, 4th, 6th, 9th) will quite radically change the amps' sound, and likely not for the better. So, distortion cancellation may actually be a bad thing in audio if not judiciously applied, although it is almost invariably a good thing in measuring equipment.
It should be noted that this is somewhat different than what happens in a PP amps where there are 'mirror images' of large portions of the amp (in the case of Baby Huey here, the whole amp is symetrical). This arrangement will supress all even order harmonics to some degree because there actually are two complementary halves, unlike the SE case where complementarity is spread into two consecutive stages that we hope to match and get a cancellation effect.
Many very prominent people in audio post here under a pseudonym. Using a real name may give more notoriety, but at least in my 'engineers' way of thinking, gives no more or less credibility than a pseudonym - what is written does.
Pseudonyms, however, do not give a clue how much 'armchair' people's discussions are. You never know what the particular persons background and experience may be, except from what is posted, possibly in other threads.
The problem I see here is precisely this notion of 'expressing negative opinions'. Are you claiming that the RHxxx amps only have positive characteristics, and only merit positive opinions on all their aspects?
Besides, I really have not seen anything in this thread tht would shed a generally negative light on RH amps, in fact just the oposite, but only opinions that just like any other construction, or in particular, form of feedback, this one is not perfect and can't be used without thinking things through hoping to automatically achieve a superb result. I really do not understand what was the problem in the first place, and why these inflamatory reactions. But now that they are over, hopefully we will return to the main topic.
Alex Kitic said:
Yet there are tons of posts on these very forums saying that the total number does not matter if you do not know the harmonic distribution. It is a simple matter looking at the amount of odd harmonics in the simulator, with a dB scale - unless you are using truly high screen resolutions, harmonics 40-50dB below the fundamental (0.5-1% distortion) will not be seen on a linear scale, regardless of the order.
If you have used something like curve captor to get a realistic model, they will show up at higher power levels. Pretty much any SE amp that I have ever seen, or at least read the relevant documentation about, exhibits this distortion - RHxxx will not be an exception. After all, where do you think the odd harmonics come from in a PP amp? They are not a magical property of the PP principle, but appear because PP is not able to cancel them out unlike even harmonics, so thay had to be made in the stages of the amp in the first place - and pretty much every PP amp can be split into two SE equivalents.
Interesting that it is you calling what I write 'mumbo jumbo' - this is a first for me and I have never done so to you. But, you have asked a question:
If single stage of amplification produced only the second harmonic, two cascaded stages will produce harmonics up to 4th, by the process of harmonic multiplication - simple, really, the 4th harmonic is the second harmonic generated in the second stage, of the second harmonic generated in the first. Similairly, if a single stage generates harmonics up to 3rd, the output will contain 2nd, 3rd, 4th, 6th, 9th. Of course, now it is a question of how you define 'higher order', and of course what amounts we are talking about, but 9th would certainly qualify.
All of this is not accounting for various methods of applying feedback, or nonlinearities of other components, mostly the output transformer.
It shouldbe noted that two inverting cascaded stages may be set-up so that inverse distortion cancellation takes place. However, it is important to know that it will cancel most of the second harmonic, but may still generate the 4th harmonic by harmonic multiplication, which is quite counter-intuitive. In theory, if the stages were completely equal, all even harmonics could cancel out, but in cases where we are speaking about a driver and output tube, this is so difficult it may as well be impossible - it may happen with a particular tube, but will not be repeatable with other examples of same tube type and probably will not hold as tubes age. Setting non-matching operating points in the square law transfer characteristic may result in lower 2nd harmonic due to cancellation, but the non-aligned curves will produce a composite transfer characteristis with a 4th order nonlinearity.
Because the human ear automatically supresses harmonics that have a monotonically decaying (on a dB scale!) amplitude with rising order, canceling out only the 2nd harmonic but leaving 4th (or 3rd, 4th, 6th, 9th) will quite radically change the amps' sound, and likely not for the better. So, distortion cancellation may actually be a bad thing in audio if not judiciously applied, although it is almost invariably a good thing in measuring equipment.
It should be noted that this is somewhat different than what happens in a PP amps where there are 'mirror images' of large portions of the amp (in the case of Baby Huey here, the whole amp is symetrical). This arrangement will supress all even order harmonics to some degree because there actually are two complementary halves, unlike the SE case where complementarity is spread into two consecutive stages that we hope to match and get a cancellation effect.
Many very prominent people in audio post here under a pseudonym. Using a real name may give more notoriety, but at least in my 'engineers' way of thinking, gives no more or less credibility than a pseudonym - what is written does.
Pseudonyms, however, do not give a clue how much 'armchair' people's discussions are. You never know what the particular persons background and experience may be, except from what is posted, possibly in other threads.
The problem I see here is precisely this notion of 'expressing negative opinions'. Are you claiming that the RHxxx amps only have positive characteristics, and only merit positive opinions on all their aspects?
Besides, I really have not seen anything in this thread tht would shed a generally negative light on RH amps, in fact just the oposite, but only opinions that just like any other construction, or in particular, form of feedback, this one is not perfect and can't be used without thinking things through hoping to automatically achieve a superb result. I really do not understand what was the problem in the first place, and why these inflamatory reactions. But now that they are over, hopefully we will return to the main topic.
gingertube said:
I think Lars has done an excellent job in presenting a problem in applying feedback this way with triodes. This is one reason why I asked about grid current in the ECC83/12AX7 at low negative grid voltages.
Lars' diagrams show that pushing the operating point to higher currents also automatically reduces the amplification factor of the stage without feedbac (steeper DC load line) which limits the amount of feedback that can be applied in the first place, keeping input sensitivity the same, or before we run into driver stage clipping.
Although linearity of the driver and amp improves, the resulting output impedance of the amp increases, making it suitable for fewer speakers. So, in the end, as always, a compromise of sorts is in order...
Wow, man, I envy you!!! You must be living in paradise!
Unfortunately, I live in a flat and lack space for everything and anything...
I've exhcanged all the CRTs in the household for LCD/plasma some time ago... so I do not have anything to fiddle with -- will have to buy something new
Thanks for the sympathetic approach!
ISSUES, ISSUES...
My father taught me that only dishonest people hide their eyes when speaking with others -- applicable to sunglasses, as well.
The only reason one would want to use a pseudonym is avoiding identification. Why would one want to avoid identification, that is a different issue.
On the other hand, my name is real and I really live where it says on the left side of this text... etc. If I were better looking, I would put my picture as "avatar". I've got nothing to be ashamed of, nor anything to hide for.
Of course I am not claiming that RH amps have only positive characteristics -- among other issues, there is nothing perfect in life or engineering. What I do claim is that if the proper attention is taken while designing (as it was) the potential problems are avoided, and advantages come forth.
It is difficult to shed negative light on the RH amps, mostly due to the fact that a lot of people have already built those, and were satisfied with them -- more than average satisfaction, I dare say. On the other hand, it is not about negative light being shed, but doubt being cast without particular reason... and not by you. Read again carefully.
When it comes to you, as previously said, we never had anything, neither quarrel nor understanding. What I do not understand is why you had to enter the discussion in the first place, saying that I dissapointed you (paraphrase of your words). As if I care.
Regarding the rest, I really will not try to reply. This thread really needs different posts than this one.
Last but not least: is anyone going to try applying the plate to plate feedback as suggested in the previously posted schematic? So far I am not aware that anyone has tried that one, and I am not interested in PP amps myself...
My father taught me that only dishonest people hide their eyes when speaking with others -- applicable to sunglasses, as well.
The only reason one would want to use a pseudonym is avoiding identification. Why would one want to avoid identification, that is a different issue.
On the other hand, my name is real and I really live where it says on the left side of this text... etc. If I were better looking, I would put my picture as "avatar". I've got nothing to be ashamed of, nor anything to hide for.
Of course I am not claiming that RH amps have only positive characteristics -- among other issues, there is nothing perfect in life or engineering. What I do claim is that if the proper attention is taken while designing (as it was) the potential problems are avoided, and advantages come forth.
It is difficult to shed negative light on the RH amps, mostly due to the fact that a lot of people have already built those, and were satisfied with them -- more than average satisfaction, I dare say. On the other hand, it is not about negative light being shed, but doubt being cast without particular reason... and not by you. Read again carefully.
When it comes to you, as previously said, we never had anything, neither quarrel nor understanding. What I do not understand is why you had to enter the discussion in the first place, saying that I dissapointed you (paraphrase of your words). As if I care.
Regarding the rest, I really will not try to reply. This thread really needs different posts than this one.
Last but not least: is anyone going to try applying the plate to plate feedback as suggested in the previously posted schematic? So far I am not aware that anyone has tried that one, and I am not interested in PP amps myself...
Guys,
I have been using the latter design fixed biased "Baby Huey" as my main HiFi Amp for some time now and have been very happy with it.
On the weekend just passed I dug out the original CCS biased "Baby Huey" (the circuit back on page 1) and did some experiments with it.
1st - In response to something posted on the "Could this be a baby Huey Killer" thread I tried a different feedback scheme. Instead of anode to anode feedback I used EL84 anode back to the 12AX7 grid on the opposite side of the push pull. This was "nice" with possibly a little more detail BUT it lost its "Baby Huey"ish ness. I did'nt like it that much and changed it back.
2nd - One of the things which worked well with the latter more complicated fixed bias design was to change the feedback such that it came of the EL84 Screens, actually the OT Ultralinear taps, rather than the anodes. "Baby Huey" meets "E-Linear" if you will. The 47K + 16K + 47K divider from the EL84 anodes was replaced with a 10K + 9K1 + 10K divider from the Ultralinear Taps. There is something magic about this arrangement. I've never heard the pace and rhythm of the amp sound better. This was my experience with the fixed bias version as well.
So - for all who have built either version of the "Baby Huey" may I RECOMMEND this mod.
1) Change the 15K or 16K or whatever X-Couple resistor you are using to 9K1
2) Change the 47K dividers to 10K and shift the connections from the EL84 anodes to the Ultralinear taps (on the same side as before).
All it costs is a few resistors.
Cheers,
Ian
I have been using the latter design fixed biased "Baby Huey" as my main HiFi Amp for some time now and have been very happy with it.
On the weekend just passed I dug out the original CCS biased "Baby Huey" (the circuit back on page 1) and did some experiments with it.
1st - In response to something posted on the "Could this be a baby Huey Killer" thread I tried a different feedback scheme. Instead of anode to anode feedback I used EL84 anode back to the 12AX7 grid on the opposite side of the push pull. This was "nice" with possibly a little more detail BUT it lost its "Baby Huey"ish ness. I did'nt like it that much and changed it back.
2nd - One of the things which worked well with the latter more complicated fixed bias design was to change the feedback such that it came of the EL84 Screens, actually the OT Ultralinear taps, rather than the anodes. "Baby Huey" meets "E-Linear" if you will. The 47K + 16K + 47K divider from the EL84 anodes was replaced with a 10K + 9K1 + 10K divider from the Ultralinear Taps. There is something magic about this arrangement. I've never heard the pace and rhythm of the amp sound better. This was my experience with the fixed bias version as well.
So - for all who have built either version of the "Baby Huey" may I RECOMMEND this mod.
1) Change the 15K or 16K or whatever X-Couple resistor you are using to 9K1
2) Change the 47K dividers to 10K and shift the connections from the EL84 anodes to the Ultralinear taps (on the same side as before).
All it costs is a few resistors.
Cheers,
Ian
I'm using some local product for the 12AX7.
Adamus,
30 years ago we had a Philips Valve manufacturing plant here in Adelaide. I believe the plant was basically a Telfunken plant. For the 12AX7 everything was made locally except that they imported the cathode assemblies from Holland. There was a wonderful story about how local guys thought the grid winding machinery was basically rubish and they redesigned and re-built it. The improved design was then re-exported to all the Philips plants around the world. I remember having a tour of the factory back when I was a 1st year apprentice in 1973.
I have a small stash of those locally produced 12AX7s - on the tube tester they show a gm of 2.2 mA/V and their side to side balance is fantastic.
I did use JJ ECC803S originally but found only 2 out of a batch of 6 I had were balanced well enough to use. I've also used ElectroHarmonix in a "BH" and found that they were generally balanced much better but I still had to select from the stock I had.
As far as using the "BH" for Rock I found that the EL84 CCS Bypass Caps are fairly critical. I'm now using 1000uF/25V Blackgates. Also don't go too high on the cross connect shunt feedback resistor. I've run up to 33K BUT like it best back at 15K. When you change to the 10K + 9K1 + 10K arrangement off the Ultralinear taps the 9K1 is chosen to give the same level of shunt feedback as 15K did with the 47K + 15K + 47K off the EL84 anodes. You can of-course adjust that 9K1 to suit as you did before. 8K2 for looser more lively sound, 10K or more for a bit tighter less dynamic sound.
Cheers,
Ian
Adamus,
30 years ago we had a Philips Valve manufacturing plant here in Adelaide. I believe the plant was basically a Telfunken plant. For the 12AX7 everything was made locally except that they imported the cathode assemblies from Holland. There was a wonderful story about how local guys thought the grid winding machinery was basically rubish and they redesigned and re-built it. The improved design was then re-exported to all the Philips plants around the world. I remember having a tour of the factory back when I was a 1st year apprentice in 1973.
I have a small stash of those locally produced 12AX7s - on the tube tester they show a gm of 2.2 mA/V and their side to side balance is fantastic.
I did use JJ ECC803S originally but found only 2 out of a batch of 6 I had were balanced well enough to use. I've also used ElectroHarmonix in a "BH" and found that they were generally balanced much better but I still had to select from the stock I had.
As far as using the "BH" for Rock I found that the EL84 CCS Bypass Caps are fairly critical. I'm now using 1000uF/25V Blackgates. Also don't go too high on the cross connect shunt feedback resistor. I've run up to 33K BUT like it best back at 15K. When you change to the 10K + 9K1 + 10K arrangement off the Ultralinear taps the 9K1 is chosen to give the same level of shunt feedback as 15K did with the 47K + 15K + 47K off the EL84 anodes. You can of-course adjust that 9K1 to suit as you did before. 8K2 for looser more lively sound, 10K or more for a bit tighter less dynamic sound.
Cheers,
Ian