[L0rdGwyn]

Well, I'm reading. You explained it very well too. The voicing of an amp seems to have a great deal to do with the behavior of its harmonic distortion profile....... this is a fascinating subject for sure.

You're referring to the spikes at 2k, 3k, 4k, 5k that are getting progressively smaller right? It forms something of a bell curve to my eyes. This is why so many push-pull amps sound dry to me, because the 2nd order is getting cancelled out but the higher order harmonics are still there.

This and the psychology behind how our subconscious mind processes sound go hand in hand for me. Both are super deep and super interesting.

Have you tried the Omegas with this configuration yet? I have a hunch that their strengths will get reinforced enough to give some wow factor there. This is why I like my Alnicos so much, because the things it does so well are entrancing enough that I can mostly just forgive the things they aren't doing so well. Mostly. There are days when I too would like to actually have some bass in my setup. There's always the TV system I guess. I have a JL Audio Dominion sub in that one just to give the towers extra omph down low that they honestly didn't even need..... but that setup is just lifeless for music playback compared to my tube amps. It's got no soul.

I definitely think the harmonic profile is a major contributor, but of course it has to be more complex than that. That article I posted is one of the few I have found that discussing the effect of harmonic distribution on the perception of frequency response changes. Nelson Pass has done quite a bit of work on figuring out how harmonics affect the subjective listening experience too. Similar to what is seen in SET amplifiers, they have found that a higher H2 and lesser H3 results in the psychoacoustic illusion of increased soundstage and space. As a result, Pass Labs specifically designs their amplifiers such that what distortion is left over (since they are still looking to achieve low distortion) is H2 and H3 in a certain ratio. While I am not a SS guy, I secretly would love to own one of his First Watt class A power amplifiers and pair it with a tube preamp of my design but I digress.

I can't say specifically that the harmonic changes going from 841 with no NFB to EF37A with NFB are the soul reason for the change in character of the amplifier, but I think it is definitely a contributor. One of the DIYers I have alluded to who is using plate-to-cathode feedback with a pentode gain stage is actually planning to implement a switch in his design that changes the pentode load from a CCS to resistive, the goal of which is to alter the sound of the amp from a high NFB typical "hifi" sound to more of a "SET" sound with a resistive load, I thought that was a pretty neat idea

As the amp stands right now, I have optimized the gain stage for maximum NFB while maintaining maximum power output of the 801A. What I think I will do next is reduce the NFB, while keeping the output impedance and FR at acceptable levels and see how it changes the subjective sound. Obviously the measurements will say "more NFB is better", so we will see if the subjective experience says otherwise.

Maybe you could try one of Omegas subwoofers? I haven't tried the Omegas with the amp yet, but you're right, they would benefit from the low output impedance.

I think tomorrow I am going to change some parts in the Maida regulator, see if I can get something near 0V bias point on the grid without changing my bias supply too dramatically, then change the OPT out for the 3.3K 60mA. As I mentioned, somehow two different builders are getting 8W out of a 3K OPT, which does not align with any loadline I can draw on the 801A curves, there must be some magic happening at this bias

 

[Zachik]

So, without question, the addition of local NFB with the EF37A in place of the 841 without negative feedback was a massively positive change, WAY exceeding my expectations.

I know how hyperbolic this all must sound, but its just blowing my mind...even compared to my 6A5G and 45 amplifiers.

I am curious whether or not you'd lose any of the magic of this new design by replacing the EF37A tubes?
I am asking, because I have a really bad experience with top cap tubes! Tried both 6C8G and EL32, and both acted as antenna for EMI. Living in a densely populated area with 20+ WiFi networks in range is not a good idea with top cap tubes. Ferrite beads on the cable to the top cap did not help. Disconnecting my router did help a lot, though...
So, do you think different / equivalent tube type that does not use top cap could work in this design?
Also, how likely your first "community friendly" amp be using this topology / architecture? or is it way too pricey due to expensive parts?

Just thinking out loud...

 

[L0rdGwyn]

I am curious whether or not you'd lose any of the magic of this new design by replacing the EF37A tubes?
I am asking, because I have a really bad experience with top cap tubes! Tried both 6C8G and EL32, and both acted as antenna for EMI. Living in a densely populated area with 20+ WiFi networks in range is not a good idea with top cap tubes. Ferrite beads on the cable to the top cap did not help. Disconnecting my router did help a lot, though...
So, do you think different / equivalent tube type that does not use top cap could work in this design?
Also, how likely your first "community friendly" amp be using this topology / architecture? or is it way too pricey due to expensive parts?

Just thinking out loud...

Hey Zachi - I haven't had any RFI issues since adding in the EF37A, they are very quiet so far. A properly positioned and valued grid stopper resistor should take care of it. I suspect using adapters in the GOTL in combination with the top cap tubes likely is a contributor, as this places the grid stopper even further from the grid input. I imagine it would be difficult to avoid RFI where you live though! Sorry you have to deal with it.

I would have to find a pentode that is similarly linear with similar gain and no top cap to work in this design, I have not looked but there might be something out there. One big advantage of the EF37A though is it is capable of taking up to a 450V B+ and g2 supply, whereas similar AF pentodes like the 6J7G are maxed out at a 300V supply. Since the EF37A can take more voltage, I can easily run it off the same power supply rail as the 801A. If I do end up moving the bias point of the 801A to closer to 320V on the plate, this would open up more driver tube possibilities.

With all of that being said, this amplifier is not a good choice for a community build. Certainly a two-stage amp using a pentode input and NFB is possible, but it is done in this design out of necessity with the 801A output tube to improve output impedance and FR. This is done for use with speakers, the NFB isn't necessary for headphones as they are typically a much higher impedance and good damping factors can be had even with the rather high output impedance of a SET amplifier. This is one reason designing a headphone amp is an easier task than designing a speaker amp this amp is also fixed bias with an A2 output stage, bias adjustment will probably be internal, so this isn't really a good one to build for other people. It is also expensive!

I am hoping / planning to put together a simple parafeed headphone amplifier that I could potentially build for others, using the Sowter 8665 OPT, possibly a single-stage "spud" design, keeping the number of active components in the signal path to a minimum with high-quality parafeed output caps. This idea is in the queue with the transformers ordered, but that is as much as I can say about it at the moment. That will be the next project after I get all of this other stuff wrapped up.

 

[Tjj226 Angel]

Hey Zachi - I haven't had any RFI issues since adding in the EF37A, they are very quiet so far. A properly positioned and valued grid stopper resistor should take care of it. I suspect using adapters in the GOTL in combination with the top cap tubes likely is a contributor, as this places the grid stopper even further from the grid input. I imagine it would be difficult to avoid RFI where you live though! Sorry you have to deal with it.

I would have to find a pentode that is similarly linear with similar gain and no top cap to work in this design, I have not looked but there might be something out there. One big advantage of the EF37A though is it is capable of taking up to a 450V B+ and g2 supply, whereas similar AF pentodes like the 6J7G are maxed out at a 300V supply. Since the EF37A can take more voltage, I can easily run it off the same power supply rail as the 801A. If I do end up moving the bias point of the 801A to closer to 320V on the plate, this would open up more driver tube possibilities.

With all of that being said, this amplifier is not a good choice for a community build. Certainly a two-stage amp using a pentode input and NFB is possible, but it is done in this design out of necessity with the 801A output tube to improve output impedance and FR. This is done for use with speakers, the NFB isn't necessary for headphones as they are typically a much higher impedance and good damping factors can be had even with the rather high output impedance of a SET amplifier. This is one reason designing a headphone amp is an easier task than designing a speaker amp this amp is also fixed bias with an A2 output stage, bias adjustment will probably be internal, so this isn't really a good one to build for other people. It is also expensive!

I am hoping / planning to put together a simple parafeed headphone amplifier that I could potentially build for others, using the Sowter 8665 OPT, possibly a single-stage "spud" design, keeping the number of active components in the signal path to a minimum with high-quality parafeed output caps. This idea is in the queue with the transformers ordered, but that is as much as I can say about it at the moment. That will be the next project after I get all of this other stuff wrapped up.

I wonder if you drop the output impedance of the 801a enough, could you get away with less output transformer impedance.

 

[L0rdGwyn]

I

I wonder if you drop the output impedance of the 801a enough, could you get away with less output transformer impedance.

Yeah exactly, I am going to try it out, maybe you missed it in one of my wall of text posts, but I am going to throw my 3.3K LL1620 60mA in the prototype with a near 0Vg bias point as a test run, they are coming out of the 6A5G amp since it is being powder coated. PB and audiowize got 8W out of an 801A A2 amp at around a 320V 60mA bias with a 3K primary. I don't see how from the load lines I am drawing, but two different people got the same result, 8W is pretty nuts...

 

[Tjj226 Angel]

I


Yeah exactly, I am going to try it out, maybe you missed it in one of my wall of text posts, but I am going to throw my 3.3K LL1620 60mA in the prototype with a near 0Vg bias point as a test run, they are coming out of the 6A5G amp since it is being powder coated. PB and audiowize got 8W out of an 801A A2 amp at around a 320V 60mA bias with a 3K primary. I don't see how from the load lines I am drawing, but two different people got the same result, 8W is pretty nuts...

I guess I must have missed it. Im a bit distracted these days to say the least lol.

 

[L0rdGwyn]

I guess I must have missed it. Im a bit distracted these days to say the least lol.

I feel ya, sometimes it feels like I have two jobs working on these amps, but it keeps my mind occupied from all of the chaos.

 

[Tjj226 Angel]

I feel ya, sometimes it feels like I have two jobs working on these amps, but it keeps my mind occupied from all of the chaos.

Yeahhhh, my truck is still in the shop, so I am just flat out useless and all my projects are piling up. I HOPEFULLY get it back in a few hours, but I have had to chase after these guys more than once, so I am not holding my breath.

 

[Zachik]

If I do end up moving the bias point of the 801A to closer to 320V on the plate, this would open up more driver tube possibilities.

I am following your thread, and will most definitely keep an eye on such development

With all of that being said, this amplifier is not a good choice for a community build.

This is done for use with speakers, the NFB isn't necessary for headphones as they are typically a much higher impedance and good damping factors can be had even with the rather high output impedance of a SET amplifier.

It is also expensive!

Does the NFB circuit make it expensive and thus not a good choice for community build? or is it expensive regardless of NFB?
For me, having speaker taps is a clear advantage. But not a showstopper if they're not there...

I am hoping / planning to put together a simple parafeed headphone amplifier that I could potentially build for others, using the Sowter 8665 OPT, possibly a single-stage "spud" design, keeping the number of active components in the signal path to a minimum with high-quality parafeed output caps.

Interesting... which tubes would you use for such community build?
Also, do you have a good pointer for reading on parafeed in general, and how it differs from OTL and SET amps? Since I own Glenn's OTL and EL3N amps, I would like my next amps to sound different from them (since I do not intend to get rid of any of them )

 

[whirlwind]

I am hoping / planning to put together a simple parafeed headphone amplifier that I could potentially build for others, using the Sowter 8665 OPT, possibly a single-stage "spud" design, keeping the number of active components in the signal path to a minimum with high-quality parafeed output caps. This idea is in the queue with the transformers ordered, but that is as much as I can say about it at the moment. That will be the next project after I get all of this other stuff wrapped up.

It will be interesting to see how this turns out...this should have a nice dark back round for headphone listening.

Many people love the "spud" designed amps.

 

[2359glenn]

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

 

[L0rdGwyn]

Yeahhhh, my truck is still in the shop, so I am just flat out useless and all my projects are piling up. I HOPEFULLY get it back in a few hours, but I have had to chase after these guys more than once, so I am not holding my breath.

Sorry to hear that, hopefully they get it up and running again, that can be a real mental burden.

Does the NFB circuit make it expensive and thus not a good choice for community build? or is it expensive regardless of NFB?
For me, having speaker taps is a clear advantage. But not a showstopper if they're not there...

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.

Interesting... which tubes would you use for such community build?
Also, do you have a good pointer for reading on parafeed in general, and how it differs from OTL and SET amps? Since I own Glenn's OTL and EL3N amps, I would like my next amps to sound different from them (since I do not intend to get rid of any of them )

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.

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

 

[L0rdGwyn]

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

I have heard great things about the EML-20B Glenn, could be a good choice with a 3.3K anode resistance with a 10K OPT primary. With a mu of 20, could get something like 150mW out of a spud headphone amp with it. Would have to be a bit larger to house filament regulators / transformers, but it is an intriguing idea, I'll add it to the list of possibilities. They are pricey though, youch.

 

[2359glenn]

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

 

[L0rdGwyn]

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

That transformer must be a custom job from Lundahl, I can't find it on their site. Parafeed has many advantages too - smaller OPT for a more compact amp, better low frequency performance with higher primary inductance, using a cascode CCS load on the output rather than a choke will give over 100dB PSRR, which means a much simpler, smaller, and cheaper power supply with very low ripple. The goal is to make a great sounding amp that is compact and cost effective, so a parafeed spud design fits well, I think the clarity will be very good with a single gain stage. I think the trick is the cap setup has to be tuned by ear, getting great results in my 45 amp taking that approach.