As above 1:3 connection (150:1.35K), one winding as primary, three others in series. Not quite 1:4 but a useful step-up. The 2P29L and 4P1L have the most gain, but 112A at mu=8.5 is a better sound. A20 seem to be easy to get used - no doubt several were made. This should give you a good sounding input stage.
Thanks, now I see, impedance vs ratio difference. Now my second question... A-20 is rated for 8ma of DC offset. If my first stage is under 8ma then can I use an A-20 wired 1:3 as an interstage? Then battery bias the second tube via the A-20 secondary? For a capacitorless coupled Amp. IOW in low current applications can A-20 be a good interstage? What if I used two A-20's in series, one at the input before the tube and another as the interstage after the tube? Giving 1:3 right at the input, then 1:3 again at the interstage. Sorry for the Newby questions, but I do own two A-20's to do this kind of experiment in mono.
Windcrest - I think you are confusing A-20, an input transformer, with the interstage transformers A-17 and A-19. A-20 does not allow any DC current.
In theory, A-20 can be used as interstage, but only with DC-blocking capacitor on the primary side. A-20 has enough inductance to work as 10K:10K interstage.
sser2:
Yes, I see I was getting them mixed up, thanks for your patience.
I've really wanted to begin experimenting with this hand drawn circuit concept that you posted last month on another thread, where you bias the 2A3 through the interstage secondary and battery biased the input tube same way. Doing some experimenting with this circuit would help me along a lot in learning some fundamentals.
sser2 All transformer 75 2A3 concept
Now for parts, I don't have a 1:2 interstage like in the drawing, but I do have a 20k:20k 20ma interstage transformer. Is a 20k:20k 20ma interstage of any use with your drawing using type 75 tube or maybe a different tube? I'm not concerned about the gain just experimenting, just want to get started using parts I have on hand.
I have the following iron, question is what can I do with a 20k interstage I guess:
A-20
20k:20k 20ma interstage
3k OPT
Sorry, It was 57, not 75. It is same as 6C6 or 6J7G, but with 2.5V heater.
57 in triode connection has amplification factor (mu) of 20. A transformer-coupled stage has gain close to mu. With 2 Vrms input from a typical CD player, a 57 driver will produce about 40 Vrms, which is more than enough to drive SE 2A3 to full power. So, 1:1 interstage transformer is perfectly fine.
20K interstage transformer impedance is also perfectly fine for the 57 tube. It's plate resistance is 10K, so when it is driving a 20K transformer, the transformer's bandwidth and maximum level will be better than specifications.
We have to apologize - 57 is an indirectly heated tube.
Size comparison of foil and metallized capacitors:
Top: 50 uF 440 VAC foil.
Bottom: 45 uF 660 VAC metallized.
9 V battery for size.
Top: 50 uF 440 VAC foil.
Bottom: 45 uF 660 VAC metallized.
9 V battery for size.
Here's the 46 input stage I'm currently using. I really like it. SUT is Hammond 1140-LN-C, 1:4, which replaces the 100K grid resistor.
Sound is very fast and clear with good treble. Particularly lifelike on jazz recordings. I was a jazz musician for 40 years and it really takes me back to being on the bandstand.
Mu of the 46 is 5.6 rather than 8 for the 10Y and 112A I've also built. Just enough for me with not much to spare. But worth it for the sound.
Coupling cap in this case is 0.1uF teflon FT-2 since it goes straight into my 2a3 outputs.
Sound is very fast and clear with good treble. Particularly lifelike on jazz recordings. I was a jazz musician for 40 years and it really takes me back to being on the bandstand.
Mu of the 46 is 5.6 rather than 8 for the 10Y and 112A I've also built. Just enough for me with not much to spare. But worth it for the sound.
Coupling cap in this case is 0.1uF teflon FT-2 since it goes straight into my 2a3 outputs.
I am looking for a SUT (configurable as 1:2 or 1:4) for levels of up to 12v (+24dBu). I would prefer to buy a good one to start, and I am thinking about the Sowter 9063. The LL7903 is more expensive and has 1:8 which I will probably never use, but allows a higher level than the Sowter.
My plan is 1.4-1.8 V rms DAC (selectable voltage) -> 01A (w/ Gyrator and SF, with divider for gain 2 - 4) - > AVC -> SUT (1:2-4) -> 49 (gyrator + SF) -> 45. I would aim for a gain of 8 for the 01A and SUT combined. If I am not mistaken, I would have (with 1.8 rms output on DAC) around 40Vpp on the grid of the 49, which should put out enough swing (+/- 160Vpp) for the 45.
I would appreciate any comments on the Sowter (which is my first option now) and on the setup described above.
My plan is 1.4-1.8 V rms DAC (selectable voltage) -> 01A (w/ Gyrator and SF, with divider for gain 2 - 4) - > AVC -> SUT (1:2-4) -> 49 (gyrator + SF) -> 45. I would aim for a gain of 8 for the 01A and SUT combined. If I am not mistaken, I would have (with 1.8 rms output on DAC) around 40Vpp on the grid of the 49, which should put out enough swing (+/- 160Vpp) for the 45.
I would appreciate any comments on the Sowter (which is my first option now) and on the setup described above.
When a mfg specifies their line input transformers as step down 2:1, 4:1. 8:1 etc. Can any of those also used as SUT by simply reversing it? For example the Monolith line level transformers are all spec'ed as step down, but I find it hard to believe that a company like Monolith with such a broad range of products doesn't make a SUT.
http://www.monolithmagnetics.com/products
http://www.monolithmagnetics.com/products
Could you put the SUT directly after the DAC? That's what I do with my Hammond 1140-LN-C which I like.
Do you need 3 tube stages? I have just DAC > SUT 1:4 > 10Y > 2a3. When I tried 3 stages the sound lacked the kind of clarity I got with 2 stages.
Hi Andy, yes, I guess if the AVC comes right after the DAC the SUT could come after it (if the AVC follows the SUT, the impedance would be too high for the AVC). If I place the SUT after the DAC and the AVC after the preamp I would have a low impedance output for the AVC but the voltage at it's input would probably be too high or at least not optimal (11.2 Vrms if I keep the preamp+SUT gain at 8 (which is what I need for the 49), using the lowest (1.4 Vrms) output of the DAC.
I am not sure if having the AVC in the first position would be so good, then, when the listening level is not very high, noise from the preamp would be more prominent.
Regarding the preamp, I don't need it (IFF using another tube as driver) but I like the sound of the 01a and it helps for the use of the AVC. While 2 stages might have benefits, I don't want to limit myself to that solution.
BTW, I am trying to learn this stuff and I appreciate your comments (and other's).
I am not sure if having the AVC in the first position would be so good, then, when the listening level is not very high, noise from the preamp would be more prominent.
Regarding the preamp, I don't need it (IFF using another tube as driver) but I like the sound of the 01a and it helps for the use of the AVC. While 2 stages might have benefits, I don't want to limit myself to that solution.
BTW, I am trying to learn this stuff and I appreciate your comments (and other's).
160Vpp is the EML45B territory, the "normal" 45 requires -about- 100Vpp grid swing in A1.
I have Sowter 9063 and like it. More better, than any 1:2 ... 1:4 SUT I tried (BTW I'm not a fan of commercial Lundahl transformers).
My 2 cents . If AVC is before SUT not only is there a high impedance but most important it cannot handle much DC offset which means a cap is needed which is a mooted point.
Hi Andy,
I have a 71a preamp that I'm currently building, but it is incredibly microphonic so far- even with rubber mounts and all kinds of counter-measures. I have it rigged with 22ma CCS and 180V B+. Sounds nice though!
Seems I can use my setup for a 112A pretty easily, since it has the same filament requirements. I just need to change the CCS to around 10ma.
- Do you find the 112A microphonic? Not sure if you can compare to the 71a.
- 180V might be a bit low, but with CCS might be ok. Thoughts?
I don't use an AVC - all my volume control is in software on my Mac Pro. So I didn't comment on where to put it. I'd assumed Dac > SUT > driver tube > AVC > output stage. I don't know the 49 but I do have a 46 driver stage which is excellent, however the mu is only 5.6 unlike 8 for the 10Y or the 112A which is very useable as a driver. The 49 has a mu of 4.7 I believe, so lower than the 46 even. With 2 stages you need to make up any loss of gain in the SUT, so while 1:4 works for me into a 10Y into a 2a3, you may be looking at 1:6 or 1:8 with the 49. Much harder to find good ones. I'll just repeat my recommendation for the 1:4 Hammond 1140-LN-C which is part of their broadcast series and very much one of their better products. It's very good value and I didn't feel the need to go to Sowter or Lundahl which cost much more.
I can see that you've put work into the 01A stage. I don't use Ale's gyrators, I like resistor loads. But is it possible that with the gyrator and SF this could drive the 45 stage without the 49? Put a 1:4 SUT before the 01A and the AVC after it. The 49 with its low gain isn't really bringing the best value to the table. As said, you have the 10Y and 112A with a gain of 8 and 2P29L and 4P1L with more than that. The 112A is better than both, but the 10Y is the king. Even the 26 is an alternative, though I find the midrange too warm for my taste. The rest of it is nice, though.
I was, like many, looking for an easier solution with 3 stages. I tried a few versions of this, like 10Y > 46 > 2a3 or 10Y > 10Y > 2a3. They lasted minutes. The loss of sound quality was just too great and too obvious and it was an instant relief to go back to the Hammond 1140-LN-C and 2 stages. I'm totally convinced that SUT > DHT driver > DHT output is the right way to do it. As soon as I tried it I knew it was right and I haven't looked back. But you do need the right DHT driver tube.
I can see that you've put work into the 01A stage. I don't use Ale's gyrators, I like resistor loads. But is it possible that with the gyrator and SF this could drive the 45 stage without the 49? Put a 1:4 SUT before the 01A and the AVC after it. The 49 with its low gain isn't really bringing the best value to the table. As said, you have the 10Y and 112A with a gain of 8 and 2P29L and 4P1L with more than that. The 112A is better than both, but the 10Y is the king. Even the 26 is an alternative, though I find the midrange too warm for my taste. The rest of it is nice, though.
I was, like many, looking for an easier solution with 3 stages. I tried a few versions of this, like 10Y > 46 > 2a3 or 10Y > 10Y > 2a3. They lasted minutes. The loss of sound quality was just too great and too obvious and it was an instant relief to go back to the Hammond 1140-LN-C and 2 stages. I'm totally convinced that SUT > DHT driver > DHT output is the right way to do it. As soon as I tried it I knew it was right and I haven't looked back. But you do need the right DHT driver tube.
I've never been troubled by microphonics with the bigger DHTs including the 112A. As in my above post I'm sold on SUT > DHT driver > DHT output. The 71A has too low gain for 2 stages, and for me 2 stages are very important. Yes - try the 112A. The 10Y is better if you can find any at a decent price. Needs Rod Coleman V9 regs and a big heatsink. The 112A is much simpler. You get the essential DHT sound without the extra finesse of the 10Y, but it's a good solution. I don't use a CCS, I use a resistor load for the sake of instrumental tone. I find a CCS compromises the tonal quality too much. I listen to a lot of classical, opera and instrumental so tone is very important. So I won't comment on a CCS. If you have 300V B+ you should be able to get about 8mA through a resistor stage, maybe more.
If you are using a 1:4 SUT you need at least 10V bias for the driver stage since a nominal 2V from a DAC will be amplified by 1:4, so the driver stage has to be carefully planned.
Thanks Andy. I'm just making a preamp for a solid state power section, so don't need the extra SUT. I can deal with the more gain of the 112A vs 71A.
I actually found the CCS to sound better to my ears vs the resistor. I tried both, but to each his own. Also, the CCS really helps counteract the crazy variance in tubes. I've found the CCS brings all the tubes to less than 1dB of each other, while with resistor loading I was finding up to 3dB. Gets expensive buying so many tubes to try to find appx matches.
I'm a bit surprised by people saying they haven't had trouble with the microphony. Maybe not everyone is measuring. It's not so noticeable by ear, but on the distortion analyzer I can easily get 20-30db of sound to register by yelling directly into the tube. If I put a speaker right next to the tube, much more. Might need to put the amp in a different location, but I'd be surprised if my experience is unique. I got my noise floor down pretty low, so trying to keep the microphony effects below the -100dbV level if possible. Not quite there.
I actually found the CCS to sound better to my ears vs the resistor. I tried both, but to each his own. Also, the CCS really helps counteract the crazy variance in tubes. I've found the CCS brings all the tubes to less than 1dB of each other, while with resistor loading I was finding up to 3dB. Gets expensive buying so many tubes to try to find appx matches.
I'm a bit surprised by people saying they haven't had trouble with the microphony. Maybe not everyone is measuring. It's not so noticeable by ear, but on the distortion analyzer I can easily get 20-30db of sound to register by yelling directly into the tube. If I put a speaker right next to the tube, much more. Might need to put the amp in a different location, but I'd be surprised if my experience is unique. I got my noise floor down pretty low, so trying to keep the microphony effects below the -100dbV level if possible. Not quite there.
9063 ordered!
yes, with 160 Vpp (which is a rough calculation) it will be more than enough for the 45, but I think the headroom won't hurt. I have not yet used the 49 and I don't really know what to expect. I wonder it it might be able to swing enough voltage for a 300B at around 4-6ma in good shape.
BTW, what is your take on using a voltage divider before the SF on the preamp?
I think I will be keeping the preamp, mainly (but not only) due to the use of the TVC. The 46, 47 and 10Y are driver options too, specially for a 300B output I also want to experiment with. I like the lower filament current demand of the 49, and those others are getting expensive, specially to 10y.