| ThSpeakerDude88 |
Cathodyne, LTP, or schmidt?
It seems a lot of amps use cathodyne, I have seen some guitar amps that use Schmidt, which is very similar to the LTP.
Cathodyne seems simplest, will it be sonicly rewarding? I don't need much gain, I am only using line level sources with my amp and not phono level. |
|
|
| Poindexter |
Opinions are split. Andy Evans likes cathodyne (concertina, split load), and I like the singly-driven diff-amp (long tailed pair). With the LTP you can sometimes lose a gain stage in your circuit, which is bound to be a plus. The 'dyne, of course, has less than unity gain.
That didn't help much, did it?
Poinz |
|
|
| Tom Bavis |
| Cathodyne has limited voltage swing - less than half supply for sure. So it's best used with output tubes that don't require a lot of grid swing like 6BQ5, 7591, 6V6... never mind what Dynaco did with it! |
|
|
| ThSpeakerDude88 |
| Id be using it for 6v6/el84 type tubes. Also noted that my 7868 amp seems to use a type of cathodyne PI. Only weird difference is that the cathode resistor is in series with one of the grid resistors on the lower op tube. I can elaborate with a schematic if needed... |
|
|
| Miles Prower |
In a case like this, it's almost a case of six-of-one, a half-dozen of the other. The cathodyne has excellent AC and harmonic balance, so long as it isn't going to be looking into grids driven positive (in that case, the unequal impedances at the plate and cathode will throw it out of balance).
I tend to prefer the LTP since I use fixed bias and cathode follower grid drivers, and the LTP eliminates one extra stage. |
|
|
| ThSpeakerDude88 |
| I looked at it again and it does appear to be a LTP. However, the resistor that normally goes from v1B to ground goes to the ground where the cathodes of the output tubes tie in, and one of the 470k grid resistors is tied to the top of the 12k resistor. |
|
|
| ThSpeakerDude88 |
I have to use a LTP or schmidt if I want to do fixed bias on a pp amp , right? it seems most amps use these in this case.
Also.. I would like to use 12au7 if possible, as I have some RCA cleartops with matched GM readings ( two tubes, four identicly matched triodes)
Any way to sub An AU7 for an AX7 with 220k plate resistors instead of 100k? |
|
|
| Miles Prower |
| quote: | Originally posted by ThSpeakerDude88
I have to use a LTP or schmidt if I want to do fixed bias on a pp amp , right? it seems most amps use these in this case. |
You don't "have to". For fixed bias, it is best to use a DC coupled driver of some sort. I prefer to use cathode followers for this purpose. However, it is also possible to DC couple to a driver that uses a triode with a good amount of g(m), like 6SN7s, or those 12AU7s. You would need an extra source of HV so that you could elevate the cathodes of the drivers above ground to establish your Q-Point bias.
The main disadvantage to the cathodyne is that it doesn't produce any gain, and so you have to separate the gain function from the phase splitting function. An LTP can do both at once, and you frewquently find that you don't require an extra gain stage.
| quote: |
Also.. I would like to use 12au7 if possible, as I have some RCA cleartops with matched GM readings ( two tubes, four identicly matched triodes)
Any way to sub An AU7 for an AX7 with 220k plate resistors instead of 100k? |
Draw a loadline and see what you get. The 12AU7 has a much lower u-Factor (~20 v. 100) and a much higher g(m) (3.1mA/V v. 1.6mA/V) and a higher Pd (2.75W v. 1200mW). It will have less gain, but higher current sourcing capability. |
|
|
| Sherman |
Though I haven't built the amp yet I now have a pair of phase splitter autoformers from Jack Elliano at Electra-Print. He has a schematic on his site for a PP EL84 using them.
My plan is to use a 6N1P for input into the PSA-2N for phase splitting and straight to the EL84s for output. A bit different from Jack's schematic but I had already designed it with the exception of the phase splitter.
I tested several other trafos for phase splitting and found some that worked really well and at low cost ($20 for the Edcors). Just something else to consider. |
|
|
| ThSpeakerDude88 |
Making a phase inverter using an interstage certainly cuts down the cost , parts count, and makes it simple. However , sound quality is tremendoulsy dependant on the transformer, so it makes it a little harder than one would think.
With PP 6BQ5's, what are the advantages/disadvantages of adding fixed bias? |
|
|
| Sherman |
Yes the trafo is key in this type of phase inverter. That's why in the end I had a couple made by Jack at Electra-Print. He has a couple specifically designed for phase splitting. They are autoformers rather than transformers. On the downside they are a more expensive solution at about $55 each.
In the tests I did the best results didn't come from the interstages I tried but from PP output trafos! The inexpensive Edcor PP trafos at $20 did a great job. I connected them without using the secondaries. I took the signal from the input tube into one of the primaries, connected the center tap (normally used for B+) to ground and then connected each primary to the grid of an output tube. Pretty easy and very good sound. The trafos are a bit big though so mounting can be an issue. |
|
|
| FlaCharlie |
Sounds interesting. What model of Edcor? What are the specs? I'm sure I've got some orphaned OTs around.
Will this work for PP using other output tubes, specifically 6B4Gs?
Thanks . . . Charlie |
|
|
| Sherman |
The Edcors I tried were XPP15-8-8K. I also tried the TF110-48-UL from Triode Electronics which also worked well. Finally I tried a potted toroidal power transformer 120v:120:12V:12V. With the primary connected it worked great with the secondary connected, not so well. Finally I tried an Edcor SE trafo with UL taps but it didn't sound very good either. (It sounds great as a regular OPT though, especially for $20.)
I suspect any PP OPT might work. Keep in mind that I was only going on sound and don't pretend to have 'golden ears'. My scope went south on me just before I started testing these trafos so I couldn't do a real frequency response test. I did compare the sound against a couple other amps I have both SS and valve.
I also tried a standard concertina type phase splitter with the same output stages. I tried these experiments with both EL84 and KT88 output stages. With the EL84 setup I used a 6N1P as the voltage amp and with the KT88s I used a 6SL7 as the voltage amp. B+ in the EL84 setup ranged from 325V to 360V and with the KT88 I tried from 350V to 400V. |
|
|
| zobsky |
| quote: | Originally posted by Sherman
Yes the trafo is key in this type of phase inverter. That's why in the end I had a couple made by Jack at Electra-Print. He has a couple specifically designed for phase splitting. They are autoformers rather than transformers. On the downside they are a more expensive solution at about $55 each.
In the tests I did the best results didn't come from the interstages I tried but from PP output trafos! The inexpensive Edcor PP trafos at $20 did a great job. I connected them without using the secondaries. I took the signal from the input tube into one of the primaries, connected the center tap (normally used for B+) to ground and then connected each primary to the grid of an output tube. Pretty easy and very good sound. The trafos are a bit big though so mounting can be an issue. |
Hi Sherman,
1. Are you saying that, for phase splitting applications, the edcor OPTs (used as autoformers) worked better than the Electra-Print PSA-2 autoformer-splitters ?
2. Which interstage did you compare with?
3. What difference did you hear comparing the "iron" splitters to the tube-based splitters.
Thanks |
|
|
| Sherman |
| quote: | Originally posted by zobsky
Hi Sherman,
1. Are you saying that, for phase splitting applications, the edcor OPTs (used as autoformers) worked better than the Electra-Print PSA-2 autoformer-splitters ?
2. Which interstage did you compare with?
3. What difference did you hear comparing the "iron" splitters to the tube-based splitters.
Thanks |
1- No, I'm not saying that at all. I tested the OPTs then ordered the purpose built Electra-Print splitters. I just haven't built the amp with the Electra-Prints yet. I tore down my bread-boarded amps before I received the PSA-2N autoformers so I haven't even tested them yet. I'll actually start building the amp next week and I'll report back when it is done (I work pretty slowly though so it could be a while.)
2- I tried three different interstages, one from Edcor- XSM10K-10K, one with no identifiable markings I got on eBay and an old Triad pulled from a console (though I don't think it was original to the amp)
3- For the EL84 setup I tried both 12AU7 and 6922 concertina splitters. They both worked fine though I might give a small edge to the 12AU7. It seemed that the bass was more defined but it was definitely not a big difference. For the KT88 setup I only used a 6SN7 for the splitter. Again it worked fine.
Bottom line, to my ears the iron splitter with the Edcor OPT was a match for the tubes. I just with I could have done some FR tests for more objective info. |
|
|
| zobsky |
| quote: | Originally posted by Sherman
1- No, I'm not saying that at all. I tested the OPTs then ordered the purpose built Electra-Print splitters. I just haven't built the amp with the Electra-Prints yet. I tore down my bread-boarded amps before I received the PSA-2N autoformers so I haven't even tested them yet. I'll actually start building the amp next week and I'll report back when it is done (I work pretty slowly though so it could be a while.)
2- I tried three different interstages, one from Edcor- XSM10K-10K, one with no identifiable markings I got on eBay and an old Triad pulled from a console (though I don't think it was original to the amp)
3- For the EL84 setup I tried both 12AU7 and 6922 concertina splitters. They both worked fine though I might give a small edge to the 12AU7. It seemed that the bass was more defined but it was definitely not a big difference. For the KT88 setup I only used a 6SN7 for the splitter. Again it worked fine.
Bottom line, to my ears the iron splitter with the Edcor OPT was a match for the tubes. I just with I could have done some FR tests for more objective info. |
Thanks for clarifying. Please report back when you finish with the PSA-2N. It should be an easy "plug and play" replacement for the edcor "autoformer" connected OPT, assuming you load the secondaries as described by Jack.
FWIW, I have a pair each of PSA-2N and conventional electra-print interstages sitting around, I hope to get started on my PP 2A3 when I can find some time. |
|
|
| fred76 |
Hi,
Looking at the electraprint schemas, it seems that Jack prefers to put a R across the PSA-2N to damp LF Resonance and 'get away' with using a low value coupling cap. I'm just don't know how one calculates for the resonance point (~5-8Hz?) using a damping R across the choke especially if one would also use a plate choke for the driver tube. Or maybe it entails one of those spice programs to get ballpark values for an initial breadboard amp. |
|
|
| zobsky |
| quote: | Originally posted by fred76
Hi,
Looking at the electraprint schemas, it seems that Jack prefers to put a R across the PSA-2N to damp LF Resonance and 'get away' with using a low value coupling cap. I'm just don't know how one calculates for the resonance point (~5-8Hz?) using a damping R across the choke especially if one would also use a plate choke for the driver tube. Or maybe it entails one of those spice programs to get ballpark values for an initial breadboard amp. |
When I spoke with him some time ago, he suggested not bothering with a driver plate choke, and just go with a plate resistor on the driver. |
|
|
| ThSpeakerDude88 |
| So is there really anything wrong with just plain using the cathodyne in the ST35? I understand its obviously not the BEST amplifier in the world, but still darned good sounding, and I am dieing to have a decent sounding tube rig running again that will at least sound better than my solid state amp. |
|
|
| SY |
| No, nothing wrong with it at all. You could drop in the input stage from Morgan Jones's Bevois Valley design or my Red Light District and you'll have a very fine-sounding amp. |
|
|
| ThSpeakerDude88 |
I have been wondering how I can rework that input stage by the way..
I know a lot of people use a 12ax7 and a 12au7, but that causes a few problems such as cross talk is greater, and the need for perfectly matched tubes to obtain true stereo balance.
I have some good matched 12au7 side "D" Ghetter cleartops, each tests @ 81/81 on my B&K 700.
How should I modify the circuit to accept these? |
|
|
| SY |
Crosstalk is close to a zero issue. The section-to-section coupling could possibly degrade the 20kHz separation by a few dB, which is not going to be audible.
The less said about using a 12AU7, the better. |
|
|
| ThSpeakerDude88 |
| quote: | Originally posted by SY
The less said about using a 12AU7, the better. |
why is that? |
|
|
| Miles Prower |
^^^^
The 12AU7 isn't the most linear triode out there. You see it used as an RF amp, or in quasi-digital applications such as multivibrators and related devices such as RS gates, or frequency dividers. It's not real good for audio, though it looks better than the 12AV7. The 12AU7, however, seems to work pretty good as an SRPP or Mu-stage, or as an LTP with active tail load. |
|
|
| SY |
| quote: | Originally posted by ThSpeakerDude88
why is that? |
Compared to similar triodes (6SN7, 6FQ7/6CG7...), the distortion is high, and high in odd-order stuff. Bleh. |
|
|
| oshifis |
| This is because the 12AU7 is a variable mu triode. Such tubes were used in AGC circuits. |
|
|
| ThSpeakerDude88 |
| would it be best to convert the sections to 12ax7's then? I do seem to see a lot of preamps however, with 12au7's in them... |
|
|
| Johan Potgieter |
SpeakerDude88,
No.
12AX7 has a too high mu and rp to make it useful for much of a driver for power tubes, etc. Also the anode does not go low enough at Vg1=0V. Try to keep to something with low rp and mu around 40-50 to avoid too much Miller cap. problems. Tubes like 12AT7, ECC88 etc. are quite good.
The 12AU7 in a pre-amp is not bad as the voltage (current) swing is low and mu not too variable over that small swing. But I would not prefer it even there.
I read a while ago that the zero gain of the cathodyne was mentioned as a factor. But if one counts that an ltp uses two triodes at half full gain, then the cathodyne actually has more gain, using the other triode as a "full" voltage amplifier, as is usual. |
|
|
| ray_moth |
| I'd avoid 12AU7 if I were you. It was not designed as a low-distortion tube for audio use and it is quite poor in that role. You would be better off with 12AT7, as Johan says, or 6SG7 or 6FQ7, which are 9-pin equivalents of the 6SN7. |
|
|
| ThSpeakerDude88 |
I didn't think of a 12AT7....And now that you mention it the 6SN7 equivalents either, as I know a lot of hi-fi amps use that as a driver/pre. I have heard of the 6FQ7 but not the 6sg7.
I would like to be able to use a tube that can do both the pre and the driver side of the 12dw7...short of just buying a 12dw7.. btw is there anything wrong with that as long as I get them matched? JJ has current production. |
|
|
| Johan Potgieter |
Nothing really wrong but ....
The 12DW7 is half 12AX7 and half 12AU7 (i.e. a "good" 12AU7). Depending on what goes where ....... Very high mu and quite low mu. Of the lot I would prefer the 6FQ7, but then really a rather higher mu tube like (again) 12AT7 or ECC88. 12AX7 can do duty in pre-amps (low microphonics), but again one must consider h.f. response; proper design.
Ray, I think you meant 6CG7 (same as 6FQ7)? The 6SG7 is a pentode, old type, more for r.f. work. |
|
|
| ray_moth |
| Johan, you're right, I meant 6CG7. Considering the availability of good alternatives in 9-pin, I am amazed that so many amp designs use the 12AU7. |
|
|
| ThSpeakerDude88 |
Johan: So basically the 12DW7 is close or like a 12au7 in one half, only with better linearity and distortion?
By the way, are there any other dissimilar triodes I might try, such as 12DT8?
I am however liking the idea of using a 6FQ7, however will it only be able to be used in one stage or two? Its too bad I'm not using 6V6 output stages, an all bottlehead amp would be pretty sweet looking. I just might have to do that sometime... 6v6's and 6sn7's. |
|
|
| Dave Cigna |
I think you have a lot of choices. If you want to copy the ST35 front end exactly you could use discrete 12AX7 and 12AU7, 1/2 of each for each channel. Of course, if you go that route then you have a mountain of alternatives to the 'AX7 and 'AU7. Maybe 5751 gain stage and 12BH7 splitter, just as an example.
If you want to stick to a single bottle per channel there are still a lot of choices. The EL84 is stupid easy to drive and the 6V6 only a little less so. Almost anything with mu greater than 15 or so would give enough voltage gain with an EL84, though you might want to double that (or more) if you intend to use negative feedback. With a 6V6 you might want to double that again. 12AT7 would work. I like 5751 better. Millions of good sounding EL83 amps have been made with a single 12AX7 for input + splitter.
If you'd like to try dissimilar triodes in a single bottle there are a few like the 6DR7. 1/2 is similar to 6SL7/5751 with mu=68. The other 1/2 is a small power triode similar to a half-size 2A3. Definitely overkill for a concertina phase splitter, but sometimes overkill is fun! :) Should have no trouble driving PP 6V6.
-- Dave |
|
|
| Johan Potgieter |
| quote: | Originally posted by ThSpeakerDude88
Johan: So basically the 12DW7 is close or like a 12au7 in one half, only with better linearity and distortion? |
Yes.
| quote: | | By the way, are there any other dissimilar triodes I might try, such as 12DT8? |
My tube manual gives the 12DT8 as an 12AT7 equivalent - not dissimilar triodes. I would still prefer that, respecting what Dave said above but 12AX7 not my choice. The 12AX7 could give stability problems as an input stage with an unknown G1 impedance (Miller effect) - not getting too technical.
If of any interest I prefer (in a one-tube format) to use a pentode-triode here (6BL8, 6U8, etc.). I simply get a better open-loop set-up, but the choice may be personal. (Pentodes are sometimes shunned as input stages because of "higher noise". That is theoretically true, but it is still inaudible unless you poke your head right into the loudspeaker.)
Apart from my small difference, good advice from Dave C. |
|
|
| Cycline3 |
Poor 12AU7.... Poor 12AU7. Always getting the rundown... I feel I have to say a little here. If I had to dog a 9 pin, it would be the 12ax7. Outside of my guitar amps, they have no place in my gear. But back to the 12au7. Somewhere down the road I came across a bunch of old stock 12au7a and 5963 tubes in like new condition. As I had lots of them, I've used them and made some great sounding pieces with them. In fact, what I am listening to right now uses them as the pre-amp and driver tubes. Check it here:
Go have a look here...
And that setup sounds very nice. It will do a Borat "crush you" on normal hifi gear out there. Ive listened to a lot of expensive systems that dont sound as swell as my little set there... Now my main amp uses 6sn7s - and I prefer them (that's another one!) but I am in the process of rebuilding it now... but to sum up - almost any tube can sound really fine... it comes down to creating the circuit for that tube that makes it work and sing. I think many are put off by the low gain, but used properly... those tubes can make some magic sounds too. |
|
|
| Johan Potgieter |
Cycline3,
Yes, the gain is uncomfortably low for many applications, and the linearity is worse than for the almost equivalent, 6SN7. Measured facts, so other alternatives simply give better designs.
Sound ... if you like it, it is your money. But as was said before, sound is subjective; some hate the same things others like. For whose taste must I as a designer design? Can't go that way - at least going by measurements I can motivate my choices. Simple as that. |
|
|
| ThSpeakerDude88 |
| I found a cleartop version of the 6FQ7 by RCA. Do these sound good? Not that being a cleartops has anything to do with it, but they sure would look sexy on the amp. |
|
|
|
