Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Creating DIY Open-Source Tube Amp Project - Input Requested
New Posts  All Forums:Forum Nav:

Creating DIY Open-Source Tube Amp Project - Input Requested - Page 3

post #31 of 128
4/2/10 at 12:50am
Thread Starter 
Quote:
Originally Posted by regal View Post
You can take a different approach, his design is a single gain stage. Look at the 6EW7 datasheet, it would be ideal as one section is a gain stage and the second a drive. So with 2 tubes stereo you get higher gain for high imp phones and lower output impedance for low impedance phones. Bottlehead's amp is similiar but he uses a slightly different tube(same family) which presents some compromises for headphones.
I'm rereading the thread to see if I missed any juicy nuggets. I assume you mean in this as 2 separate class A stages where the second gate is fed from the plate output of the first? I just want to make sure I am understanding you completely =)
Reply
post #32 of 128
4/2/10 at 2:06am
Quote:
Originally Posted by dBs View Post
I'm rereading the thread to see if I missed any juicy nuggets. I assume you mean in this as 2 separate class A stages where the second gate is fed from the plate output of the first? I just want to make sure I am understanding you completely =)
Yes but the single tube contains the tube very different stages. I have a schematic at work I can foward you, it is for s ingle-feed but parafeed probably makes more sense because you can CCS feed the output stage and it doesn't reuire such a demanding powersupply.

Right now I am building a SE OPT 5842 amp with Salas HV shunt PS, pratcically no caps in the audio cicuit.
Reply
post #33 of 128
4/2/10 at 4:00am
You should look into the PSR of Mu followers. I la-la-love to hear that Im wrong about these thigns, but PSR on a Mu follower is no where near as good as CCS loaded plate output.

Assuming Im actually right about that:
You may be better off finding the path to lower output impedance elsewhere with better PSR. What does another active stage cost to add to the circuit with a different turns ratio on the OPT? What does the PS quiet enough to meet your standards of a quiet output with a mu-follower cost? compare them.
Reply
post #34 of 128
4/2/10 at 6:55am
This basically takes the best features of the Bottlehead S.E.X amp and the Woo6 and combines them for an amp with good gain for headphones, good PSRR, relaticely inexpensive. I decided to go the single 5842 route instead because I want to avoid caps altogether, but this option I considered is much more versatile ( the 5842 amp I'm building is a Grado only thing.)

This is just a rough scetch of what I was trying to describe in words to give you some thing to think about. I know you seem to be leaning against SET amps but they are the way to go, being an EE if you could really crack the code on a quiet background DHT SET you would have the low distortion you seem to be after and an amp that would be up there with the most expensive options. How hard can it be to make a quiet musical .25A 5V filament supply for a recent EE grad who hasn't forgot all the book stuff?



Edit: Mistake 2nd tube after CCS should be 150V at the plate, negative bias at the grid should read 17V. Sorry but I will fix the schematic when I get a chance.
Reply
post #35 of 128
4/2/10 at 11:49am
Quote:
Originally Posted by nikongod View Post
You should look into the PSR of Mu followers. I la-la-love to hear that Im wrong about these thigns, but PSR on a Mu follower is no where near as good as CCS loaded plate output.
I think you are spot on, Mr. nikongod.

The mu follower does have less power supply ripple rejection, yet allows the current in the lower tube to stay more-or-less constant. Meanwhile the CCS as an anode load has very large power supply rejection, yet allows the current in the lower tube to vary as current is transferred to the load.

So it's a case of weighing up the theoretical pros and cons: mu-follower has lowest distortion owing to near-horizontal load line yet requires more intensive power supply treatment, while CCS anode load requires less power supply investment yet has higher distortion owing to non-horizontal load line.

That's something for dBs to consider.



Quote:
Originally Posted by nikongod View Post
Assuming Im actually right about that:

You may be better off finding the path to lower output impedance elsewhere with better PSR. What does another active stage cost to add to the circuit with a different turns ratio on the OPT? What does the PS quiet enough to meet your standards of a quiet output with a mu-follower cost? compare them.
You are right about that. However the psu for the mu-follower could be a TL783 for each channel (plus a few resistors and caps). Not sooo expensive, but certainly more complicated to build.

Horses for courses. As you say... "compare them".


Have a nice day, one and all.
Reply
post #36 of 128
4/3/10 at 12:16am
Thread Starter 
Everyone is bringing good ideas to the table.

Naturally I don't have the wealth of experience that you all seem to have, so it would be wise of me to ask if traditionally the PSU or the amp itself is the more difficult element to design and balance (balance the trade-offs)? I imagine that the answer will be "it depends on if you try to design it without ______ or with ______". If I were to go for a mu follower and have to consider a more perfected--complex?--PSU, would that difficulty only be marginally more so than the one demanded by a CCS? Conversely, if the CCS is a more difficult amp to design, is that a marginal increase in complexity versus the mu follower? Which increase in difficulty/complexity the bigger hurtle, PSU or amp?

If that didn't make sense, a distinct possibility, than I can clarify.

I like the ideas floating around now though. There's still a fair bit of time yet before I have the means to begin purchasing parts.
Reply
post #37 of 128
4/3/10 at 1:06am
Well, there be more than one way to skin a cat, especially in DIY land. In my day job (electrical engineer, not electronics engineer) I recommend that folk fully document the specification then start the system design by considering the end and then work their way back. (Obviously it takes some wiggle and adjustment along the way, and it’s not an entirely linear process).

Anyway, that would translate in your task here as: choose an output transformer, design the output stage to drive it, (then if you are going to have one) design the input stage to drive the output stage, then design whatever supply was needed by the stages in order to achieve their objectives.

Or something like that.
Reply
post #38 of 128
4/4/10 at 7:27am
dBs, I don't know if you have seen this, but it might be adaptable to your needs: http://www.forsselltech.com/download...VT%20Opamp.PDF
Reply
post #39 of 128
4/5/10 at 12:56am
Thread Starter 
Thank you for sharing that. Actually, that confirms pretty well what I had been thinking along the lines of. I have a biiig white board (best things on Earth) where I put up ideas, sketches, quick notes, etc. and my sketch looks pretty similar to that. I haven't involved any differential pair and there are a few things I'm not entirely certain as to the reason for (though I'm sure there are good reasons), but I do like that there are more similarities than differences =D

I'm starting my research a bit now into the power supply end of things. I understand this area better than I do amp design since my classes were more relevant in this arena. I do have some general questions though; mostly with grounding. Grounding is one area that can cause major problems with ground loops, improper grounding, chassis vs earth, isolated grounds, etc.. This is one area where I think colleges really need to almost dedicate a class on as it is pertinent to almost all areas of electrical engineering.

I am also curious as to the advantages of dual power supplies as is used in my 337. I would see that as a risk for ground loop problems, PSU and part variability, and susceptibilities. It isn't something I plan on putting into this particular design for obvious reasons, but I am still curious.

Quote:
Originally Posted by regal View Post
This basically takes the best features of the Bottlehead S.E.X amp and the Woo6 and combines them for an amp with good gain for headphones, good PSRR, relaticely inexpensive. I decided to go the single 5842 route instead because I want to avoid caps altogether, but this option I considered is much more versatile ( the 5842 amp I'm building is a Grado only thing.)

This is just a rough scetch of what I was trying to describe in words to give you some thing to think about. I know you seem to be leaning against SET amps but they are the way to go, being an EE if you could really crack the code on a quiet background DHT SET you would have the low distortion you seem to be after and an amp that would be up there with the most expensive options. How hard can it be to make a quiet musical .25A 5V filament supply for a recent EE grad who hasn't forgot all the book stuff?
Your design definitely looks unique. I wont take it just by virtue of the fact that it wouldn't be mine and I don't to steal your idea. You did remind me of the possibility of using diodes off of the cathode though. I read very briefly about that a while ago and was curious, but have forgotten to investigate further. I only ran across it that one time so it is a seldom used technique.

Were you planning on using some kind of PRIMM or SS designed current source?

As far as being able to crack DHT hum problems, the easiest thing that comes to mind is using a separate power supply for the heating. I would be surprised if that hasn't been done before though so I suspect there are significant drawbacks to that method. Even though I have a piece of paper on my wall, I put more value in the experienced knowledge of people like yourself and other longtime (aged? XD) DIY designers. I only know theory. The real world is a different place, and while my ego would want me to claim I could crack that code, my sensibilities suspect that if others haven't cracked that code yet the odds aren't in my favor...yet. A few years of real world experience and I may be ready for that challenge. I will look into it despite though, it is best to know ones enemy after all!
Reply
post #40 of 128
4/6/10 at 2:16am
Quote:
Originally Posted by regal View Post
How hard can it be to make a quiet musical .25A 5V filament supply for a recent EE grad who hasn't forgot all the book stuff?

This would be really useful and really helpful to anyone that works with DHT's. Don't mean to insult you, but it would also be something specific and concrete you could really sink your teeth into, rather than flopping around and trying to "reinvent the wheel" with vague "blue sky" amp ruminations.

All DHT's require seperate filament supplies. In most cases, DC is reuired. All the usual methods (battery, brute force, volt. reg., current reg., etc.) have their own problems. Probably the best, but the least used, is ultrasonic AC. Something like a Colpitts oscillator running at 50khz into a small broadband amp might be a possibility. You could probably even make money selling it in board or kit form. I'd buy it!
Reply
post #41 of 128
4/6/10 at 3:06am
Quote:
Originally Posted by FrankCooter View Post
This would be really useful and really helpful to anyone that works with DHT's. Don't mean to insult you, but it would also be something specific and concrete you could really sink your teeth into, rather than flopping around and trying to "reinvent the wheel" with vague "blue sky" amp ruminations.

All DHT's require seperate filament supplies. In most cases, DC is reuired. All the usual methods (battery, brute force, volt. reg., current reg., etc.) have their own problems. Probably the best, but the least used, is ultrasonic AC. Something like a Colpitts oscillator running at 50khz into a small broadband amp might be a possibility. You could probably even make money selling it in board or kit form. I'd buy it!
10-4, to me this is the biggest missing link for the DIY tube headphone amp builder. DHTs practically took over the speaker amp world with their linearity and wonderful sound, yet they have barely made an impact on headphone amps because of this issue.

Pmillet has an incomplete ultra-sonic oscillator schematic for DHT filament supplies on his web page that you may be able to use as a starting point.
Reply
post #42 of 128
4/6/10 at 3:42am
Just seen the 6EW7 datasheet from RCA, and it looks like a nice tube for the job:
http://www.scottbecker.net/tube/sheets/049/6/6EW7.pdf

dBs there is another example here: 6EW7 Se amplifier

Regal, you might want to reconsider your ouput tube operating point!
Reply
post #43 of 128
4/6/10 at 3:54am
Quote:
Originally Posted by Leny View Post

Regal, you might want to reconsider your ouput tube operating point!
Its OK, max disspapation for section 2 is 10W if I posted the right schematic should be 150V*.045A= 6.75W (note the negative grid bias used to avoid the cathode cap). You want to run this thing hot to get the lowest Rp (see the last page of the datasheet).
Reply
post #44 of 128
4/6/10 at 4:01am
I meant the way it's drawn shows 265V anode to cathode (i.e. supply) and -117V grid to cathode (i.e. bias), which is not feasible.

Quote:
Originally Posted by regal View Post
... if I posted the right schematic should be 150V*.045A= 6.75W
Yup, 150V and 45 mA sounds / looks good.

: )
Reply
post #45 of 128
4/6/10 at 4:08am
Quote:
Originally Posted by Leny View Post
I meant the way it's drawn shows 265V anode to cathode (i.e. supply) and -117V grid to cathode (i.e. bias), which is not feasible.



Yup, 150V and 45 mA sounds / looks good.

: )
Yea I need to clean that up should be 150V on the plate after the CCS and -17 on the grid, originally it was DC coupled. But with the parafeed it makes more sense to try and eliminate as many big caps as possible.
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Creating DIY Open-Source Tube Amp Project - Input Requested