DIY Audio Projects Forum
http://diyaudioprojects.com/Forum/

6J6 / 6AQ5A Canned Ham Mixer SE Amplifier Help
http://diyaudioprojects.com/Forum/viewtopic.php?f=9&t=6006
Page 1 of 1

Author:  Chi [ 20 Aug 2017, 15:35 ]
Post subject:  6J6 / 6AQ5A Canned Ham Mixer SE Amplifier Help

Hi Everyone,

I'm putting together a little amplifier for my little Canned Ham Travel Trailer. So there isn't a lot of room. I have a bunch of 7 pin sockets and loads of 6AQ5 tubes, hence this direction. The source will be either a smart phone or and ipod and only one output speaker, again due to space and not a lot of need for volume in the little trailer. So i put together a schematic and I have the amp on the test bed. I just put some plate to plate feed back in and that seamed to make things sound nice. But I thought before i put this in the cute little box I'm putting together I'd ask everyone to help me tweak it.

Thanks In advance

Bob

Author:  Suncalc [ 20 Aug 2017, 22:00 ]
Post subject:  Re: 6J6 / 6AQ5A Canned Ham Mixer SE Amplifier Help

Chi wrote:
help me tweak it
There are a few things.

Your mixer triode is biased almost at conduction. Even small signals will drive major distortion. This is why the feedback helped so much. I would bias your mixer stage closer to -2v. This would mean about 870Ω/2 (due to the shared cathode) or about 430Ω. A standard 440Ω would be fine here. With this in place you can probably dump the feedback entirely.

The power stage is at an ok bias point. I would dump the 100Ω resistor on the screen. It is not required and will only serve to hurt the dynamic responsiveness of the amp. I would also increase the power stage bypass cap to about 470µf. Otherwise your low end will suffer. I would also increase the grid resistor on the power stage to 470kΩ. This is just to reduce loading on your mixer/driver stage.

Just my quick thoughts on the schematic.

Author:  Suncalc [ 21 Aug 2017, 15:09 ]
Post subject:  Re: 6J6 / 6AQ5A Canned Ham Mixer SE Amplifier Help

I have attached a load line for the mixer.
Attachment:
6J6 Mixer Load Line.jpg
The blue lines are my suggested biasing and the red lines represent the 50Ω bias in your design.

I understand that you probably took the 50Ω value directly out of the data sheet but there is something important to keep in mind. This is an RF mixer tube and, as such, the design points given in the data sheets will be good for RF design, not necessarily line level audio. As you can see the 50Ω bias puts the bias voltage down around 160mv. This is fine for low level RF but less than optimal for an audio amp. I suggest you try the updated mixer bias point with and without feedback and see what you think.

Author:  Chi [ 22 Aug 2017, 21:23 ]
Post subject:  Re: 6J6 / 6AQ5A Canned Ham Mixer SE Amplifier Help

Hi Matt,

You were right I did. I worked out the load line this morning using the numbers you gave me in your first post and you were spot on. I see in your new post I got it right this time. I understand where I goofed now , so thanks. However, is it usually considered best practice for less distortion to get the load line to be as perpendicular as possible to the plate curves by either adjusting the b+ voltage or your plate resistor, or both?

Also, I got a chance to make the adjustments you outlined with really wonderful results. The amp really sounds much better. Better bass, and even more usable gain. I did find it alittle, I guess you might discribe as brassy, at higher volumes, so I put in a small amount of plate feedback, 1 meg, and that made the performance nicer with a very small sacrifice in overall volume. Btw, my wife likes it too, so you got two happy campers with your help.

I'll make the corrections on the plan an post it in case anyone is interested.

Thanks again Matt.

Bob

Author:  Suncalc [ 23 Aug 2017, 22:01 ]
Post subject:  Re: 6J6 / 6AQ5A Canned Ham Mixer SE Amplifier Help

Chi wrote:
However, is it usually considered best practice for less distortion to get the load line to be as perpendicular as possible to the plate curves by either adjusting the b+ voltage or your plate resistor, or both?
Actually, no. This is a commonly held misconception due to making (somewhat specious) arguments about the plate characteristics and implying some importance to the way load line intersects the family of characteristics.

Although the family of plate characteristics contain all the data necessary for (at least theoretically) an infinite number of load lines, Once a load line is draw, what really matters (at least from a graphical perspective) is the dynamic transfer characteristic for that particular load line. The more linear the Dynamic Transfer Characteristic, the lower the distortion. The introduction of additional resistance in the plate circuit, causes the first term of the plate current series to dominate the higher order terms, and the dynamic transfer characteristic to straighten. So, the bottom line is the greater the plate load (all else being equal) the lower the distortion.

This can be best seen in these two examples of triode dynamic transfer characteristics.
Attachment:
Dynamic_Transfer_Characteristics.png
The figure on the left shows how the dynamic transfer characteristic changes for changes in plate load at constant grid bias and plate voltage. The figure on the right shows how the dynamic transfer characteristic changes for changes in plate load at constant plate supply voltage. In both these case it is easy to see that increasing the plate load resistance actually straightens the transfer characteristic (i.e. decreases distortion).

This is not an easy concept to grasp without delving into the math, but the graphs do help. If you still doubt, it is a simple matter to look at two triode load lines against the same set of plate characteristics using the same B+ and a middling bias resistance. The load line with the higher plate load resistance will always have the lower distortion number in class A1 operation (while not in conduction or cutoff).

I hope this helps.

Author:  Chi [ 24 Aug 2017, 06:52 ]
Post subject:  Re: 6J6 / 6AQ5A Canned Ham Mixer SE Amplifier Help

Hi Matt,

Yes. It helps very much. I also read your 6V6 SE bias optimization article which was really great. It looks like if I go to 250 volts I can get some really nice results out of my little 6AQ5. Couple that with revisiting the load lines on the 6J6 I think I have to hit the drawing board this evening! Can't wait, should be fun.

Thanks

Bob

Author:  Chi [ 31 Aug 2017, 15:31 ]
Post subject:  Re: 6J6 / 6AQ5A Canned Ham Mixer SE Amplifier Help

Okay, for those interested, the final schematic, sounds great on the bench, now building a case for it. I will post a pic when i have it done.

Bob

Attachment:
revised6j6cannedham.png

Page 1 of 1 All times are UTC - 6 hours [ DST ]
Powered by phpBB® Forum Software © phpBB Group
http://www.phpbb.com/