Jump to content

Another custom tube amp


Recommended Posts

I didn't realize until recently that there's interest in seeing some of the custom amp designs that I get myself into! The locals tell me that if I spend too much time doing only restoration work, I'll get too rusty. They hope I'll live a long time and plan on keeping me very busy.........

Attached are pix of the latest bit of insanity “commissioned” by one of the guys. Although he listens at sensible levels in a moderately sized room (with CWs, among others), he wants the ability to drive some vintage speakers, like AR3s or Large Advents, in addition. The latter eliminated a flea power SET as a possibility. He also likes a very potent bottom end. So, after lots of discussion and some demonstration, I decided to go with a parallel SEP arrangement since I don't like push-pull amplifiers. This unit can use any of the 6AQ5 family of tubes (which I consider to be the best sounding of the 6V6 equivalents, and are better sounding than the 6V6 itself to me.) He likes a very large soundstage with maximum channel separation, so dual mono was the way to go to keep channel crosstalk as low as possible. Normally, I would have built the power supply on a separate chassis and strapped it over to the amp via an umbilical cord (this provides the quietest operation since the AC powerline garbage is isolated as well as keeping the magnetic field from the power xfmr away from the output xfmrs.) He prefers a single chassis, even if it has to be large (it is- 17” X 10”), so that's how I arranged it. The only common elements are the power cord, on/off switch, and the suppression cap across the power xfmr primaries. The suppression cap not only filters a little of the noise coming in on the AC line, but also eliminates any clicks/pops when the amp is turned on and off. Power output is around 7 wpc which is sufficient for this application. One last inclusion is the variable low pass filter for use when over processed recordings cause the “ear bleed” phenomenon which I've mentioned in other posts. Sorry the pix aren't more clear. My camera is an “antique” Sony Mavica which stores a whopping 3-4 pix on a floppy!!! Can you guys tell that I'm a bit behind the times?

Maynard

post-40520-0-76200000-1393882772_thumb.j

post-40520-0-29960000-1393882788_thumb.j

  • Like 1
Link to comment
Share on other sites

very nice layout and construction Maynard. Have you ever used the 6CG7 in place of the 6SN7? I am also a fan of the 6AQ5. SEP's have great bass, how sensitive is your circuit with respect to pair matching and drift over time? Thanks for posting. Best regards Moray James.

PS: thought you might find this interesting

http://www.audiophonics.com/

Link to comment
Share on other sites

very nice layout and construction Maynard. Have you ever used the 6CG7 in place of the 6SN7? I am also a fan of the 6AQ5. SEP's have great bass, how sensitive is your circuit with respect to pair matching and drift over time? Thanks for posting. Best regards Moray James.

PS: thought you might find this interesting

http://www.audiophonics.com/

Actually, I don't use any medium-mu triodes in the front end due to their high drive requirements. I prefer that all of my amps be capable of direct drive by the source component so that a preamp won't be required. I employ either high-mu triodes like the 6SF5 used here or, if even more voltage is needed to drive the output stage, a pentode like the 6SJ7. What's nice about the high-mu triodes is that they can be contact (aka grid leak) biased which gives them all the advantages of both fixed bias and cathode bias. It allows the cathode resistor and bypass cap to be eliminated and also allows the tube to self-regulate over its lifespan. Regarding matching of the output tubes, I don't bother! I simply select a pair which measures about the same on an old emissions type tester. Operating voltages are kept at, or slightly below, the design-center rating, so if one tube draws a couple of milliamps more at idle than the other it really doesn't matter. Also, I set up the filament string so that the rated voltage isn't exceeded under a worst case line voltage situation (around here it's about 122 maximum). Most of the time, the filament voltage runs a few % low which makes a huge difference in terms of longevity. I've never encountered a problem with paralleled tubes over even many thousands of hours of use. And thanks for posting that site. It's one that I haven't seen before and her premises look interesting!

Maynard

Edited by tube fanatic
  • Like 1
Link to comment
Share on other sites

As usual, nice work!

I have a stash of 6AQ5, and a few 6005s as well. Made a guitar amp of a champ circuit using a 6AQ5/6005, nice and creamy sounding...

What primary impedance is the 125SE output transformer, and is there high-mu triode that could be subbed for the 6SF5? Though I wouldn't doubt I have 6SF5s lying about...

Link to comment
Share on other sites

As usual, nice work!

I have a stash of 6AQ5, and a few 6005s as well. Made a guitar amp of a champ circuit using a 6AQ5/6005, nice and creamy sounding...

What primary impedance is the 125SE output transformer, and is there high-mu triode that could be subbed for the 6SF5? Though I wouldn't doubt I have 6SF5s lying about...

Thanks for the favorable comments everyone! Mike, the tubes which are in the amp right now are 6005s because I had a bunch on hand. The 125FSE is wired for a 2.5k load to the tubes which, if used singly, would require 5k. I'm maxed out for the rated bias through the opt (89 ma on one channel, and 91 on the other), but based on measurements published by George Anderson on one of the others in the series, it seems that Hammond builds in a 10+% margin without having to worry about core saturation. If I had needed to buy the xfmrs, I would have gone with the 125GSE to have more of a comfort zone. The 6SF5 has no direct substitute, but the 6AV6 is electrically identical. The only concern I'd have with the latter is if it's too close to the power xfmr there may be some hum issues if you don't put a shield on it. The 6SC7 is another good choice which, if you use a separate tube for each channel, would allow you to parallel both sections to double up on the output voltage if you need it. If you have never tried contact bias, give it a shot. Ground the cathode, and put 10 meg on the grid to ground (you must AC couple the grid however) and the tube will make use of its internal space charge around the grid to generate the needed bias. As mentioned, it will self-adjust throughout its life which seems to be "forever" if the filament voltage is controlled.

Maynard

  • Like 1
Link to comment
Share on other sites

Very nice, Maynard!

I was thinking parallel single-ended when you mentioned the speakers being used. Very nicely done, those suppressor caps really work well - Dynaco used something similar across power switches decades ago, and they're obviously still very effective. I paralleling tube sections like that can be really nice -- I've done that with 6SN7 and 6SL7 driver stages with excellent results.

So, ground the cathode, 10meg grid-leak resistor, AC coupled grid. It might be a circuit where a very high impedance (AC) grid choke would work well. I prefer chokes to grid-leak resistors anyway...sort of a personal (ahem) bias.

Link to comment
Share on other sites

Here's a question for you...

How difficult would it be to make a 'stereo' amp where it's actually two mono amps on a single chassis.

Think for biamping

Ballsy amp on the bottom for bass0

Lighter amp on top for the midrange & treble

One chassis, one plug but otherwise, two amps.

Oh, and XLR inputs

Link to comment
Share on other sites

If I may, since this is a general question. Sure it can be done. Decware does that with their Mini Torii amplifier - an outstanding amp at that, I have to say. There are different ways of doing it, from splitting B+ plate supplies into two voltage rails, power transformer capability being one consideration, to using separate filament supplies, as well as dual mono volume controls to aid in adjusting gain and balance for each channel.

Or true dual mono: Using two completely separate power supplies and DC filter sections for each channel, along with dedicated input and driver stages.

Edited by erik2A3
Link to comment
Share on other sites

So, ground the cathode, 10meg grid-leak resistor, AC coupled grid. It might be a circuit where a very high impedance (AC) grid choke would work well. I prefer chokes to grid-leak resistors anyway...sort of a personal (ahem) bias.

Hi Erik! Nice to see you on the forum again. I've been a fan of grid leak bias "forever." No frequency response issues and only a couple of 10 cent resistors (plate load and grid leak). What's nice also is that you can run virtually any plate supply up to 300V and the tube will adjust its own bias accordingly- more voltage, more output and no worries about having to re-do the load line. And the distortion is very similar to that obtained with cathode bias too. The lifespan of the tubes used this way is astronomical. Even in the old AA5 table radios, it was rare for the 12SQ7 or 12AV6 to fail. It was always one of the other tubes.

Coytee, as Erik pointed out it is very feasible to go dual mono on one chassis. This amp is totally dual mono right down to the separate power xfmrs for each channel, as well as separate volume controls. To economize, you can use a really beefy output xfmr for the bass amp and one with a more limited frequency response for everything else. Or, you can do what I did in this one and use an RC filter for the voltage feedback which reduces the amount of signal fed back in the bass range. It results in a perceived bass boost which has to be heard to be believed! It's a real "cheap and dirty" method of doing this, but it sure works extremely well. Lots of ways to accomplish the desired goal.

Maynard

  • Like 1
Link to comment
Share on other sites

Baldwin built an organ amp many have seen here modified for full-frequency, two channel stereo use. Though both channels were very similar as far as input and phase-splitter, there were some subtle differences in cathode bias of the output stage, as well as output transformers used.

Maynard, I've seen your grid bias before now that I've had a look at a few schematics. I am very interested in trying it just so I can compare the sonic effects using the same input stage and tubes with more common cathode bias, plus cathode resistor bypass cap. I've never used grid-bias, and have tended to want to avoid the use of a series capacitor between source input and the grid of the first stage. I've done that as a precautionary measure, though, on a couple of designs that were otherwise completely direct coupled between input and output. All of this is ancient stuff! Vintage audio electronics that is as valid today as it was 60 and 70 years ago. Fascinating! thank you for sharing this recent work you're doing. I would think anyone working with hollow state gear would be very curious.

Link to comment
Share on other sites

Maynard, I've seen your grid bias before now that I've had a look at a few schematics. I am very interested in trying it just so I can compare the sonic effects using the same input stage and tubes with more common cathode bias, plus cathode resistor bypass cap. I've never used grid-bias, and have tended to want to avoid the use of a series capacitor between source input and the grid of the first stage. I've done that as a precautionary measure, though, on a couple of designs that were otherwise completely direct coupled between input and output. All of this is ancient stuff! Vintage audio electronics that is as valid today as it was 60 and 70 years ago. Fascinating! thank you for sharing this recent work you're doing. I would think anyone working with hollow state gear would be very curious.

Erik, I'm sure you will be very pleased with the result. Be sure to use a cap with a high enough value to prevent any low frequency attenuation. In my typical arrangement, I run the input across a 500k volume pot and go to the voltage amp grid through at least .01 uf. I have always liked AC coupling as it eliminates any concerns about DC offset or leakage from the source component. Sonically, I have never heard a difference between this method and direct coupling. And, yes, we old time radio/tv repair guys generally know all kinds of neat tricks to employ; and, as you stated, they are as valid today as they were when I first got started! I'll see if I have a copy of an article which I wrote long ago for a friend's web site which describes the construction of a simple tube amp using a variant of a vintage table radio audio stage. Of course, it was totally upgraded to offer phenomenal audio quality. Many have been built over the years both by myself and others, and never a single complaint about the sound it offers. This stuff is truly a passion!

Maynard

Link to comment
Share on other sites

I found some early versions of the article/schematic which I mentioned to Erik above. They are pdf/odt format which I tried putting on here, but the format isn't supported. So, if anyone is interested, send me a PM with your email and I'll gladly pass it along-- Maynard

Link to comment
Share on other sites

I found some early versions of the article/schematic which I mentioned to Erik above. They are pdf/odt format which I tried putting on here, but the format isn't supported. So, if anyone is interested, send me a PM with your email and I'll gladly pass it along-- Maynard

PDF documents work, no problem. If they are Open Document format, they may not allow it.

Sending you a pm...

Bruce

Link to comment
Share on other sites

I can empathize about accumulating paper and literature Bruce! The amount I have stashed is frightening at times! Thanks for the comment about pdf format being compatible here. The schematics were already in that format, but the article was in odt which I converted to pdf. I just gave everything a quick scan and things look in order.

Comments/criticisms are welcome! The article was written for the benefit of experimenters/hobbyists who wanted to try their hand at building a tube amp, and the explanations of what is going on in the amp are as clear as I could make them without using any math.

If anyone decides to build the amp, feel free to contact me with any questions or concerns. Better to ask before doing anything than try to fix things up afterwards.

Maynard

amp article.pdf

stereo amp v2.pdf

separate-ps-for-stereo-amp-1.pdf

  • Like 2
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...