[Tjj226 Angel]

Why not just use the recommended transformer and forget about parafeed leave out that capacitor no capacitor really sounds good.

There is always a capacitor in your signal path. In a normal configuration the power supply capacitor and bypass capacitor are in series. In a parafeed configuration your parafeed cap and bypass capacitor are in series.

At least with parafeed you get the noise floor reduction

 

[Zachik]

I think a spud amp using a EML-20B should sound fantastic. Always wanted to build a amp with this tube.
Will be a costly build but few parts the less parts the better the sound. KISS
http://www.emissionlabs.com/datasheets/20b-ota-1.pdf

Quick search yields more than $500 per tube!!! Yikes.

 

[johnjen]

And with a parafeed cap, what with it being a smaller value (≈5-6µfd), the suitability and availability of quality caps is much greater than power supply caps.
And yeah 'tuning it' by ear is really the only way to dial it in.
Even though high end non-polar caps in this range can be mighty expensive, at least the 'electrolytic smear' can be avoided.

JJ

 

[Zachik]

The NFB circuit is not expensive, it is mostly related to the tubes themselves (the 801A is getting increasingly rare and expensive) and the sheer number of parts, this is a pretty BIG amplifier which will have a separate PS chassis, it has six transformers total. Also, it is a fixed bias amp which is a little less user friendly than a "auto bias" amplifier, as the bias points of the tubes needs to be manually adjusted. Since this is sort of a personal statement build, the plan right now is to have the bias adjustment on the inside. Getting it set up to be accessible on the exterior with meters and pots present a totally new layout challenge and again, more parts and cost. I hope that makes sense! Just not a very user friendly design, overall, it's sort of a crazy experimental type of build.

OK. You got me convinced... not a good option for a community build, for several reasons...

Not so sure on the tubes yet, but the Sowter 8665 has a 10K primary, which leaves open a lot of possibilities. Max power output for that transformer is 500mW, so assuming I keep a 10K primary (depends on the secondary winding and impedance of the headphones used), that would mean using a tube that can swing a maximum 200Vpp into a 10K load, which means a gain of 35 maximum with a typical DAC 2Vrms input. But 500mW is overkill, a lower gain tube would be fine too, there is some flexibility, my very vague plan right now is to try a bunch of different tubes out and see what sounds best! Since I am NFB happy at the moment, maybe I will also experiment with a power pentode with cathode feedback its just going to be a sandbox until I find a good sound.

Sounds interesting. Being a single stage amp would probably make it sound quite different from my GOTL and 6EL3N amps. Right? and single stage would make it very low noise floor, I assume?
Do you plan on implementing parafeed for this amp?

Here is a resource that goes over what a parafeed amplifier is: http://diy.ecpaudio.com/p/parafeed-tutorial.html

Interesting read! Thanks.

 

[2359glenn]

Quick search yields more than $500 per tube!!! Yikes.

Not that unreasonable I paid allot more then that for 300Bs.
Might have to do something about that cathode resistor bypass capacitor and get it out of there.
My direct coupled 300B amp has no capacitors at all in the signal path except for the last capacitor in the power supply. That is a hard one to get rid of.

 

[L0rdGwyn]

Sounds interesting. Being a single stage amp would probably make it sound quite different from my GOTL and 6EL3N amps. Right? and single stage would make it very low noise floor, I assume?
Do you plan on implementing parafeed for this amp?

Yes, I would expect it to sound different! Tube choice will play a large role in that, as will the parafeed capacitor configuration. Does your 6EL3N amp have the ability to use a single stage like the GEL3N? I know Glenn included that on some other amps, that switch essentially changes the amp to a spud amp, but not parafeed.

The noise floor should be very low yes. I think they discuss it in that parafeed article, but you have two choices for the DC plate load of the output tube - a plate choke, or a CCS load.

The advantage of a plate choke is it allows the output tube to swing higher than your applied B+ voltage through energy stored in the choke (inductance). The disadvantage is they are large, expensive, must have a suitabily high inductance to allow for passage of the lowest frequencies of the audio band, they also do not have the same level of PSRR (power supply rejection ratio) as the a CCS load. You can think of PSRR as the degree to which power supply noise is squashed on the output signal.

The advantage of a CCS load is it is smaller, cheaper, gives much better PSRR to the point that the power supply does not need a significant amount of passive filtering, so it can be made very simple, small, and cheap! The disadvantages (because there are always compromises ) is the CCS has to dissipate a significant amount of heat (the reason I have heat sinks mounted to the back of my 45 amp), you can use a pentode CCS to get this heat out of the chassis, but performance is compromised. Also, there is no energy stored in a CCS allowing the tube to swing higher than the B+, meaning to get a full output swing, the amp requires a much higher B+, approximately equal to:

(output tube plate voltage + cathode voltage) + 1/2*(peak-to-peak output tube voltage swing) + (CCS dropout voltage) + (headroom).

For my 45 amp that I am reworking for speaker use, for example, I am looking at a 275V bias point on the plate. But to get the full output swing with a CCS load, I will need something like a 500V B+!

 

[L0rdGwyn]

There is always a capacitor in your signal path. In a normal configuration the power supply capacitor and bypass capacitor are in series. In a parafeed configuration your parafeed cap and bypass capacitor are in series.

At least with parafeed you get the noise floor reduction

Not that unreasonable I paid allot more then that for 300Bs.
Might have to do something about that cathode resistor bypass capacitor and get it out of there.
My direct coupled 300B amp has no capacitors at all in the signal path except for the last capacitor in the power supply. That is a hard one to get rid of.

Could use a Maida regulator and fixed bias. This is what I am doing in the 801A amp, there are no capacitors in the output stage signal loop. There is now a regulator, but the output impedance across the audio band is less than the resistance in the wires connecting it to the tube plate. There are film caps in the signal of the driver stage though, no electrolytics! Oh and there is a SiC transistor

In the future, I plan to bring this amp to a meet, find the tube purist and let them listen. Afterward I will tell them there is a transistor in the signal path and NFB, oh the horror!

 

[Zachik]

Does your 6EL3N amp have the ability to use a single stage like the GEL3N? I know Glenn included that on some other amps, that switch essentially changes the amp to a spud amp, but not parafeed.

Yes. Using this 1-stage mode, the amp becomes MUCH lower gain (obviously!), and even lower noise floor (pretty low vs. GOTL to begin with...). Work well for easy to drive dynamic driver headphones. Most of the time - I use the 2-stage mode, though.

 

[Propagandist]

This entire thread is just a wealth of wonderful, helpful information. Thank you everyone, and especially LG, for being so descriptive and detailed in your posts. As someone trying to design my own amp, your thoughts are extremely valuable.

 

[L0rdGwyn]

This entire thread is just a wealth of wonderful, helpful information. Thank you everyone, and especially LG, for being so descriptive and detailed in your posts. As someone trying to design my own amp, your thoughts are extremely valuable.

Thanks for following along! Happy to answer questions. @Tjj226 Angel @2359glenn and @A2029 helped me get started, along with a handful of other DIYers on other forums, it has been a bottomless pit ever since.

 

[Xcalibur255]

I've brought up the 20B idea off and on for years now. Basically just the same concept as the Inspire IHA-1 with a big fancy DHT in place of those 6SN7s. The IHA-1 can't even call itself a true OTA because it has each section of the 6SN7 to work with.

Yamamoto also does their C3m single tube headphone amp. Aside from that I can't think of many examples of this type of headphone amp off the top of my head.

I'm guessing the list of tubes that work well for this application are going to be pretty limited since you need both high enough gain and low enough Rp to get decent performance out of the OPT and it seems like those things don't often go hand in hand.

 

[Xcalibur255]

Not that unreasonable I paid allot more then that for 300Bs.
Might have to do something about that cathode resistor bypass capacitor and get it out of there.
My direct coupled 300B amp has no capacitors at all in the signal path except for the last capacitor in the power supply. That is a hard one to get rid of.

This is the one with the 10 driver stage right? I still want to own an integrated single chassis version of this before I die. I have rolled idea after idea around in my head about what my "ultimate" end game amp would be, and time and time again I keep coming back to 10Y >>> 300B.

AKA difficult and expensive.

 

[L0rdGwyn]

So there seems to have been a disconnect as far as what 8W out of the 801A in A2 actually means...WARNING: wall of text.

This morning, I tried out the 3.3K OPT setup with the Lundahl LL1620 60mA.

I nixed my bias supply and just used 520K to ground, not ideal since the grid voltage isn't fine-tuneable, it is fixed by the gate-to-source voltage of my FET buffer, so -3.1V on the grid of the 801A. I was able to adjust my bias point to some degree using the trimpot on the Maida regulator. I dialed it in to a 335Va / 57mA / -3.1Vg bias point.

With this bias, I am getting 156V to 524V peak-to-peak swing on the 801A plate without clipping, hard clipping on the positive peaks starts beyond 524V. This results in roughly 16V peak-to-peak into 8ohms, about 4W out. This is pretty darn consistent with the datasheet if I draw out the load line, copper losses included.

Ideally, I would get to a true 0V bias point and 320V with a 3K (rather than 3.3K) primary impedance, it is doable, have to put +3V on the gate of my FET. At that bias, if the datasheet is to be believed - and so far has been consistent with my measurements - even assuming no copper losses, that makes 5.5W into a 3K load without clipping. Even if you did something crazy and biased at +10V on the grid at 280V on the plate, that is 400Vpp into 3K, makes a 6.4W.

Here is the 3.3K / 335V / 57mA / -3.1Vg load line. As you can see, consistent with my measurements, the amp should clip at roughly 520-530V on the positive peaks.



So, I was very confused as I could not see how you could get 8W out of an 801A A2 amp at any 0V bias point. Thought maybe some magic would happen if I tried it out, but it doesn't look like it. As it turns out, the DIYers who suggested it were quoting power output from the full peak-to-peak voltage swing on the load line above, even though the 801A will hard clip on the positive peaks. So, swinging from 70V to 524V into a 3.3K load, that does add up to about 8W. However, the postivie signal is going to be chopped off above 524V.

That just isn't how I go about measuring power output, maybe my approach is flawed, but for me the power output is measured at peak-to-peak swing without hard clipping, so this is a 4W amplifier whereas I was getting closer to 5W with the LL9202 and my old 365V / 50mA / -10Vg bias point.

Just to see this experiment all the way through, I added additional NFB given the reduced gain needed from the driver stage, here are the measurements, FR curve below. I believe I am hitting transformer limitations again, as the HF rolloff is nearly identical to what I am getting in the 6A5G amplifier.

Output Z: 3.4ohms
THD @ 1W 1.52%
-0.86 db at 20kHz



How about the sound going from the LL9202 50mA with a 6.5K 100H primary to the LL1620 60mA with a 3.3K 60H primary?

While the sound is still very good, my initial impression was that it took a step back, that degree of hyper-realism, airiness, and staging was lost to some extent. I am going to give it another try later after a break.

So, while the bandwidth may have improved a tad, the output impedance is higher, we have higher distortion, more NFB, less power, and overall I feel the sound has degraded somewhat. I am likely to return to the LL9202 and rebias the 801A at 350V for a full 5W output unclipped.

 

[L0rdGwyn]

I've brought up the 20B idea off and on for years now. Basically just the same concept as the Inspire IHA-1 with a big fancy DHT in place of those 6SN7s. The IHA-1 can't even call itself a true OTA because it has each section of the 6SN7 to work with.

Yamamoto also does their C3m single tube headphone amp. Aside from that I can't think of many examples of this type of headphone amp off the top of my head.

I'm guessing the list of tubes that work well for this application are going to be pretty limited since you need both high enough gain and low enough Rp to get decent performance out of the OPT and it seems like those things don't often go hand in hand.

You are exactly right @Xcalibur255 , the ideal spud tube is high gain, low plate resistance. That is why the EL3N is a very good choice! With a 3K plate resistance and a mu of 20, it makes for a good single stage tube.

There is something I have thought to experiment with though - it just an idea now, have not even simulated it yet - but using a cascode CCS plate load on the parafeed output, you could potentially take the mu output of the CCS as opposed to the plate, lowering the output impedance significantly. This might make it possible to use higher Rp tubes in a spud parafeed design and get a low output impedance to adequately damp headphones. If it works, could open up more possibilities with high Rp / mu tubes. Maybe I will throw together another 700V PS and try it out with the 841, Rp of 25K, mu of 30 just for funsies though, not sending a 700V B+ amplifier out into the world!

 

[Tjj226 Angel]

I've brought up the 20B idea off and on for years now. Basically just the same concept as the Inspire IHA-1 with a big fancy DHT in place of those 6SN7s. The IHA-1 can't even call itself a true OTA because it has each section of the 6SN7 to work with.

Yamamoto also does their C3m single tube headphone amp. Aside from that I can't think of many examples of this type of headphone amp off the top of my head.

I'm guessing the list of tubes that work well for this application are going to be pretty limited since you need both high enough gain and low enough Rp to get decent performance out of the OPT and it seems like those things don't often go hand in hand.

The list of tubes you can do this with is actually quite large. It just all depends on what you want and how you want to do it.

First step is to figure out what headphones you are trying to drive. Lets look at the HD600/HD650 headphones as an example. I will skip the math and the technicalities and just sort of skip straight to the answers so that you can better understand what is going on without getting confused.

A HD600 has a sensitivity of 105 db at 1 volt. In order to hit 120db peaks, we need the amplifier to provide 6v peak to peak with headroom, so we will probably want to design the amp so that it can output 7v peak to peak. We know that ohms law tells us that the current we need at 7v into a 300ohm headphone is 23.333... ma. Lets just say we need 30ma as a fudge factor.

Since most dacs these days will output 2v RMS which is ~2.83 v, we need an amp with a total gain of about 2.5ish. Again, lets call it 3 for the sake of having plenty of margin to play with.

Now you need to find a combination of a tube and output transformer that will yield a total gain of 3 with a bias current of 30ma (you actually don't need 30ma for a number of reasons, but as you know by know, I like have a ton of margin in my designs).

Keep in mind that a 5K to 300ohm transformer has a voltage ratio of 4, and a 10K to 300ohm transformer has a ratio of about 6. So we need to look at tubes with a gain of somewhere in the mid teen range. A pro tip here is to choose a twin triode tube that can be paralleled. I normally don't like paralleling tubes since it is kind of a pain to find twin triodes that are well matched, but paralleling tubes will actually reduce their self noise by 3db. Considering the low winding ratio of the transformer, we need all the help we can get in order to make the amp quiet.

The ever venerable 5687 would be the perfect tube for this application. In parallel the bias current would be above 40ma, the gain is about 17, and the plate impedance would be reduced to about 1K.

I would use two united SIC Jfets as a CCS for the plate load (about. 15 bucks per channel), and a edcor xsm 15K to 600ohm transformer that you would be using as a 7.5K to 300 ohm transformer which would cost another 15 bucks per side.

Alternatively, you could use tubes like the 6E5P with an unbypassed cathode resistor to reduce gain and increase plate impedance to match the same configuration as the 5687 for even less money.