[GWorlDofSPACE]

Quote:

Originally Posted by Parafeed /img/forum/go_quote.gif Nikongod has a point. Your preferences are what matter, not ours. I already stated a preference for transformer designs. With a $1K budget both an OTL or transformer amp can be designed. Whether anything is designed as a collective or not, your preferences should guide you. Have you heard a single-ended transformer (plate) coupled amp without feedback? Have you heard something like the MAD EAR+, a single-ended transformer coupled amp, but the transformer output is coupled to the cathode of the output tube, instead of the anode - cathode follower transformer output? Do you have preferences on feedback? Global feedback is a complete no-no? You don't like the sound from cathode followers? You'd rather not have more than 6db of global feedback, otherwise you feel that you might as well switch to a cathode follower output and have 100% local feedback rather than 20db global feedback? The point is, I can tell you what I like and dislike, what I think is the best way to go, but your ears are what matters, not mine. Before putting a grand on the table to buy the parts on a list, I'd like to have some idea if a finished design is going to suit my preferences. You should too.

I can only tell you what end results I prefer. My experiences are really small. I have heard a few anode-follower DIY amps for speakers but this is some time ago. At moment I would say my Bijou as a OTL is sounding great in the mids. I would prefer a sound signature like that but more dynamic presence in the lows and highs. There needs to be also more micro details present. If I could help? I am here to learn!!!

 

[regal]

OK I'm going to throw this one out there. Haven't built it but it will be my next project after my 5842 Shunt reg Spud amp. This one just looks incredible, probably Parafeed has built it. All battery powered SE DHT, man it looks like the ultimate tube head amp:


Puppy Power 2

 

[nikongod]

Its nice to see the circuits from the radios from the early 1900's getting attention. I bet they sound nice with modern components.

Why not just build the amp on a bread board and decide what filament heating technique you like for yourself?

Since that amp already has a coupling cap why not grid bias the second stage? This was commonly done, although it did require at a minimum another battery so it was not always done in the cheap radios. You could compare that to the various cathode bias techniques in the prototyping phase too.

Why build with batteries? 1 more thing to try on the breadboard.

 

[dsavitsk]

Quote:

Originally Posted by regal /img/forum/go_quote.gif All battery powered SE DHT, man it looks like the ultimate tube head amp:

Quiz of the day: What is the output impedance of that amp, assuming the OPT ratio is 5K:32?

Bonus question: What is the ratio of the tube's load to rp? And what should it be?

 

[regal]

Quote:

Originally Posted by dsavitsk /img/forum/go_quote.gif Quiz of the day: What is the output impedance of that amp, assuming the OPT ratio is 5K:32? Bonus question: What is the ratio of the tube's load to rp? And what should it be?

output impedance 1800rp x32/5000+dcr~12-15 ohms
load to Rp = 2.7

He designed the amp for speakers to get most power with a sacrifice in distortion, obviously a different OPT would be needed for headphones. Probably 7.5-10k/32. But the idea of a battery powered DHT headamp with a black background on Grados is intriguing, no?


On the opposite end of the spectrum my Stacker II with its "SS OP transformer" has an output impedance of 1 ohm, and a load to Rp of around 10k.

 

[dsavitsk]

Quote:

Originally Posted by regal /img/forum/go_quote.gif output impedance 1800rp x32/5000+dcr~12-15 ohms load to Rp = 2.7

Effective rp of the tube is rp + (mu * Rk unbypassed), so ~8K.

Zout is the transformed impedance, which is ~51 ohms, plus copper resistance, which is typically going to be ~6 to 8 ohms on a 32 ohm tap. So, Zout is close to 60 ohms here.

 

[Leny]

Here we go...

For a project to be successful many things must be objectively defined. As this is not a commercial project the definitions are relatively simple, and include:

> Overall objective
> Cost
> Scope and limitations

From this a Technical Specification should be developed.

Only then can alternative technical approaches be assessed.


The opening poster (GWorlDofSPACE) is clearly well meaning, but instantly hampered their own thread by inferring that 'state of the art' must mean vacuum tube. The reality of such a project is that alternative technologies can not be assessed in depth until the Technical Specification is set.

If you guys are serious then take the first few pages of this thread as both an interesting Prologue and a good example of how not to start a project.

Then start again… !

……………………………………………………………………………………………………………

So, I’ll take a first shot…


> Overall objective
- To design a state of the art headphone amplifier for dynamic headphones
- Headphone impedances to range from 30 to 600 Ohms
- Output to be sufficient to power headphones with a sensitivity rating of 94 dB at 1mW to a short-term peak SPL of at least 112 dB
- There should be at least one unbalanced input and one balanced input
- There should be TRS jack, 2 x 3 pole XLR, and 1 x 4 pole XLR output connectors


> Cost
- Purchase price target for the components, excluding the enclosure, to be $1400 / £900 / Euro 1000


> Scope and limitations
- This is a DIY project
- No ‘surface mount’ electronic components
- Only specify the use of a pcb if a member has the time to design it properly, otherwise specify point-to-point
- Electrostatic headphones are not accommodated in the design as they have a different method of operation which is outside the scope of the project
- K1000 headphones are not accommodated in the design as they are of very low efficiency and hence require a large voltage swing which is outside the scope of the project

……………………………………………………………………………………………………………

Obviously this is just a first stab and very much a personal opinion. This needs to be discussed by all, so over to you…

 

[pabbi1]

Quote:

Originally Posted by Leny /img/forum/go_quote.gif > Cost - Purchase price target for the components, excluding the enclosure, to be $600 / £900 / Euro 1000

Uh, $1360 / £900 / €1000 ? If your rate was right, both the UK and the Continent would be seriously overrun with Yanks.

Sorry, I think the first two pages were just varying opinions on how such a noble goal would end with miserable fail.

 

[Leny]

Quote:

Originally Posted by pabbi1 /img/forum/go_quote.gif Uh, $1360 / £900 / €1000 ? If your rate was right, both the UK and the Continent would be seriously overrun with Yanks.

Hello pabbi1.

Yup, I typed that at 3am in the morning! I actually went back and corrected 600 to 1400 just before you posted. D'oh.

So it's nearly 4am and I'm off to bed. Hope some folk read my previous post and discuss the 'Overall objective'...

 

[nikongod]

I think it really boils down to the fact that there are simply too many ways to go about this and as mentioned elsewhere everyone has their own preconceived notions of which is the best.

The best solution is for a a single person to head the project from day 1, and throw out his/her own ideas and improve them through recommendations & advice from the community. Other threads following that model have resulted in really nice amps.

This is already a fairly hot thread. If you want to get anywhere start a new one asking for help on a basic design that you think shows promise or interests you in some way. Trying to get even 6 people to agree on what to build and start from scratch is a recipe for disaster. 4 of the 6 will get fed up with why the other 5 dont do it their way and leave. The other 2 will act like they agree until one kills the other and eats him. Which leaves you with 5 angry people and 1 leading the project. The above method has 1 person leading the project from day 1, and nobody angry.

Do you see pete millet's original hybrid headphone amp starting this kind of ****-storm thread? Nope, pete said here it is, enjoy it! The original was actually quite nifty (the TO220 buf-634 are awesometoast), The original head-fi MH was modified and improved by the community, version 2 pimped out some more, and then I lost count. But there was no serious fighting and the amp got better and better. Someone took control of the project, layed down some specific objectives and groundwork, and lead it on an excellent path for this amp to evolve and improve.



As a point of note:
Electrostatics could be accommodated with a push-pull tube amp with an output transformer to the dynamics. The cost can be quite low (IMO, I did it for about $50 on top of an amp I already had) if you make a point of using "inexpensive but certainly decent" parts. Anyone who wants is welcome to upgrade

Depending on your listening level and opinions WRT how much voltage swing is required to drive 'stats you may have strong opinions about why this wont work but it is doable on the cheap if you find yourself with the right gear.

 

[regal]

Quote:

Originally Posted by dsavitsk /img/forum/go_quote.gif Effective rp of the tube is rp + (mu * Rk unbypassed), so ~8K. Zout is the transformed impedance, which is ~51 ohms, plus copper resistance, which is typically going to be ~6 to 8 ohms on a 32 ohm tap. So, Zout is close to 60 ohms here.

I guess I assumed he forgot the bypass caps on the schematic, doubt he is driving speakers without them.


Agree With NikonGod there will never be agreement, but I think we would all love to hear a DHT headamp all powered by battery.

 

[dsavitsk]

Quote:

Originally Posted by regal /img/forum/go_quote.gif I guess I assumed he forgot the bypass caps on the schematic, doubt he is driving speakers without them.

I doubt that. A lot of people think the way to get rid of that bypass is to just leave it out. But you know I was just giving you a hard time

Quote:

Originally Posted by regal /img/forum/go_quote.gif but I think we would all love to hear a DHT headamp all powered by battery.

Batteries aren't a cure all and come with their own issues. But, a neat trick is that 9V batteries will mechanically connect together in series to make a HT supply.

 

[regal]

I guess I've been drooling over Frank Cooters 71A design so much that I didn't even notice the lack of bypass caps, but you don't miss a thing!

I thought it fascinating that these old DHT's were originally designed to run on batteries for the filaments.

 

[Parafeed]

Quote:

Originally Posted by regal /img/forum/go_quote.gif This one just looks incredible, probably Parafeed has built it.

No, but eons ago I did build and use a Welborne Ultrapath pre for a time. That powered both the filaments and HT (if you can call it that) from lead acid batteries, used the ECC86 and had a tap for headphones on the Electraprint output transformer. It sounded pretty good with Senns, but not really enough power for them.

 

[regal]

Quote:

Originally Posted by Parafeed /img/forum/go_quote.gif No, but eons ago I did build and use a Welborne Ultrapath pre for a time. That powered both the filaments and HT (if you can call it that) from lead acid batteries, used the ECC86 and had a tap for headphones on the Electraprint output transformer. It sounded pretty good with Senns, but not really enough power for them.

Thanks what are your thoughts on battery powered DC filaments on DHTs like the 71a, is there still an issue of low AC filament impedance and the audio signal picking noise up from them.