|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
66.188.193.13
In Reply to: RE: Looking to build a 6V6 amp! Thoughts? posted by xaudiomanx on August 15, 2014 at 07:59:27
If you use pentode mode, the 6V6 has rather high distortion - 8% to 12%, according to the RCA specs. That pretty much means that feed back is a must. I haven't found any specs on distortion in triode mode, so that is an unknown. The plate resistance is triode mode (1960 Ohms) is reasonably low, however.
A DHT triode (45, 46, or 2A3) will give you better performance. That said, 45s are crazy expensive, 46 are becoming more popular and therefore more expensive, and 2A3s are approaching the cost of mid grade 300Bs. I'm in the design stage for a flea power headphone amp, and am toying with the idea of using a 12B4A, which has an Rp of 1030 Ohms.
Good luck with your project.
Follow Ups:
DHT's. Just want a simple amp. Other suggestions would be nice!
O.K. Getting back to your original question, I usually separate the power supply from the amp using separate chassis, and interconnect them with a cable of less then 3 feet in length. Unless you can find a magnetically shielded power transformer, the fields around it can couple into other components (e.g., output and interstage transformers). I've done it both ways, but prefer separate chassis. I built a stereo amp with separate power transformers\chokes for each channel on a single large chassis. It weighed a ton and got really hot.
The speakers you settle on have an ENORMOUS influence on everything else.
I gave you some ideas for a no GNFB triode wired design. It will sound fine, but 2 WPC is all there is. If you employ speakers whose sensitivity is in the 94 to 96 dB. range, look to a full pentode setup that uses regulated g2 B+ and employs both Schade and global NFB. The bulk of the NFB should be in the short, Schade, loop. Only a few dB. will be needed in the outer, global, loop.
Whenever GNFB is employed, use a 1 pole high pass filter that "corners" in the 15 to 18 Hz. range, to gain protection against O/P trafo core saturation. Suppress infrasonic trash! Also, employ high gm small signal types. High gm provides resistance to HF error correction signal induced slew limiting.
Eli D.
I built a push-pull 6V6 ultralinear amp with used Dynaco Z-565 outputs and a single, current-sourced, long-tail differential 6SL7 driver. I used some 0.22uF Facon paper-in-oil coupling capacitors.
There's barely enough excess gain to provide any negative feedback. The amplifier has a nice midrange but sounds pretty shut-in. I might try changing to poly coupling caps to see if it opens up a bit. The amp makes about 11 Watts per channel.
This was a junkbox project and I've never tried to optimize it. It was supposed to be a loose adaptation of Poindexter's "Musical Machine."
IMHO, not really a success. Listenable, but not extraordinary. I know, not what you were looking for, but sometimes it's good to have a data point, even if it's not a positive one.
I think adding another stage of gain would lower the distortion and give a more incisive sound. If you're going to run almost pure open-loop, you probably want a very linear output stage. And a more robust driver tube.
LOL.
-Henry
Henry, that's a very clean layout. A shame to see it sitting idle just because the first rendition didn't wake up. Have you made basic measurements like frequency response and the like? Much as I like octals, I believe 6BQ5s produce better dynamics than 6V6 in PP designs. Can't prove that, it's just my gut feel after years of hearing both.
--------------------------
Buy Chinese. Bury freedom.
Thanks. The amps haven't gone to waste. I sold them to my friend, Bryan who was using them in his system which is notably more high-end than mine. About a year ago I brought my new solid-state amplifier over to his house and it sounded so much better I took pity on him and gave it to him on semi-permanent loan. I use the tube amps to watch movies on my TV.
The amps aren't horrible, they just don't have that "sparkle" or "see-through transparency" of the solid-state amp.
-Henry
Mikey
The 12B4A are nice tubes but
6AH4 and 6EM7 are better sounding.
DanL
Short of #45 DHTs, triode wired 6V6 family tubes are the best bet in this power range, for GOOD sound. Obviously, highly efficient speakers are needed, when approx. 2 WPC is the power yield.
A 12AV7 or 5965 will provide all the drive needed in a RC coupled setup.
If the OP includes substantial amounts of reservoir capacitance on the amp's chassis, along with using a shielded umbilical between the chassis pair, I don't have objections to separating the amp from the power "iron", rectifiers, and some filtering elements.
BTW, a good performing, low cost, B+ PSU for the project, can be built around a Triad N-68X isolation transformer and a pair of Cree C3D02060F 600 PIV Schottky diodes. "Full wave" doubler topology would be employed working into very substantial stack capacitance, a "hash" filter, an "ordinary" filter choke, and reservoir capacitance.
Eli D.
I agree with Eli. 6V6 plate curves look great in triode connected.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: