My DIY 45 tube amp makes gorgeous music!!

[mg16]

I just yesterday finished my first tube amp DIY project, and damned if it didn't work exactly as claimed.
It is "Single Ended Glory" by Eric Barbour , published several years ago in Glass Audio Magazine. It is 2 watts of gorgeous single ended sound. My Lascalas love it!
I would unhesitatingly reccomend this project to anyone looking for 45 single ended amps.
Total cost was under $600. The output iron is One Electron UBT-2's. The chokes and power trannies are Hammond. Most of the parts were bought at either Handmade electronics, or Antique Electronics Supply, both on the web.
Resistors are Vishay/Dale and Mills power resistors. Small caps are mostly Sprague orange drops. Output coupling caps are MIT Multicap. All are polyproplyne film and foil except for some large paralled mylar values in the power supply. Input and loudspeaker connectors are Cardas .Sockets are white ceramic with gold contacts. Used 45's bought off of ebay for $50.00, with a $20.00 Sylvania 6sn7gtb.
It is built in a Hammond steel 17 by 10 by 3" high chassis, with all the wiring point to point , except for the 6sn7 driver and circuitry which I etched a small pcb for to simplify construction.
The only exposed parts on top are the tubes, main power trannie, and the UBT-2's.
Not much to look at, but it sounds wonderfull. I was expecting a great midrange, with mediocre bass and soft treble. What I got was a 3d midrange, plenty of bass "whomp" at the low end, and clear extended high frequencys. Very wide soundstage, and it should get even better as it breaks in.
So far, it exceeds my expectations easily.
It isn't that great to look at,(I'll fix that later by painting the exposed screw heads on top flat black, and by building a nice oak wood base for it), but last night Jimi Hendrix played "Voodoo Child" in my living room.
mg16

 

[GRollins]

The Sovtek 6SN7s are actually quite good and are reasonably priced. When you get a little further along, consider replacing the Orange Drops with something else. I use Orange Drops for experimental purposes when I'm doing R&D, but pull them out and replace them with better caps when it's time to come off the bench and into the listening room. My Orange Drops have probably seen duty in ten or twenty different circuits over the years, but have never made it to the system. Like the old saying: Always a bridesmaid, never a bride. Poor things!

Grey

 

[PassFan]

Grey:
What kind of caps do you use to decouple the stages and were do you buy them? Thanks

 

[mg16]

45 Tube amp Coupling caps

Barbour specified coupling caps between the driver and power tube stages as: 400volt, 2uf., polypropolyne . He gave no recomendations as to brand.
I used Multicap brand, 10 section, metalized polypropylyne . 2uf/400volt. I bought them from Handmade Electronics web site, along with most of the other caps and resistors, power trannie, chokes, connectors, driver and rectifier tubes, etc.
They seem to work well, and sound fine.

 

[dice45]

Coupling caps recommendation

Hello mg16, hello all,

1st, Congrats to your amp being alive!


A year ago, I made a big coupling cap listening comparison because I was fed up with 2nd hand info, I wanted to try it out myself. The test was aiming at finding out more about long-term listening properties and so I played any cap for hours, listening up to 9 critical records. The respective cap was running as coupling cap in a solidstate preamp with 100mV across it. TME, caps do not love such tiny voltage differences, they love to have some 10 or 100 Volts across them; so this was a worst-case test.

The undisputed winner was a Siemens/Epcos MKV 25839, together with the no longer made Siemens MKY. Close to them were no-brand polystyrenes (in a yellow housing with superior inner contacts , not the usual trasparent stuff) and 0.5µF silver micas. 3rd winners with sweet but a bit mute treble and beefy low end ands overall nice sound were Audyn/SCR tin foil polypropylens. 4th were different Roederstein polyprops: KP1836 and MKP 1841.

Listening to those cpas was never shorter than 2 hours uninterrupted; the MKV and MKY yielded 4 hours each and the test was hard to stop after that.

I was not pleased with Hovland MusiCap (boring!! and eating up downward dynamic range) and different Jensen paper-in-oil (be they with alu or copper, they too ate up downward dynamic range, but atleast in a pleasant way), and Teflon caps with plastic housing (same drawbacks) undisputed looser in this section was the Hovland.

All those caps mentioned above were non-magnetic. Those caps being magnetic at the leads or worse, at the contact face, or even worse at the whole housing were not only boring, the added some nastiness to the sound I cannot stand. Negative spiritual sensations, so-to-speak, they make me escape ASAP and I stopped listening to those caps very soon, after 20 to 45 minutes.

Among those test loosers were:

Teflon-Mil hermetic
Shizuki
Taitsu
Sprague OrangeDrops
Sprague VitaminQ

I find it remarkable that those caps having high praise in different audiophile DIY scenes were beaten by a Roederstein "Eroid" cap from the late 50ies having any flaw a cap can have. Among some of my friends they are called "Heino"-caps as they help to a sound like from a 50ies juke box

Someone I am corresonding with had the MIT caps and I send him MKV samples; he clearly preferred them because of the same criteria I preferred them. But this is 2nd hand info.

I would recommend the MKV cap to anyone playing around with re-capping; DUnno where to get them internationally, ask your electronic distributor. Worst case they can be ordered at www.buerklin.de or www.schuricht.de . If this doesnot work, email me, maybe I can help.

Greets,

 

[PassFan]

Wow, thanks for all the info on this subject guys. Really above and beyond the call.

 

[mg16]

Coupling caps

Dice45,
Thanks for the insight on differing brands of caps. Later on down the road, I may try some of your suggestions. I would be interested to see if the sound could get even better on my 45 amp. Right now I am happy as all get out with it, but the "tweak" bug never fails to kick in eventually, no matter the project I build!
mg16

 

[GRollins]

I used the Multicap 'styrene & foil critters in my big tube amps and am very pleased with the sound. It's been years since I've done a shootout between caps, so I do not claim that they are the best. In particular, I have not heard the Multicap polypropylene at all. I have a friend that swears that the Multicap Teflon caps are sufficient to start a new religion, and though I've heard his system, there are too many variables, and I wasn't there when he put them in in order to play back and forth and get a comparison.
And the cost...holy smokes...the cost...

Grey

 

[mg16]

Coupling caps

All I know is that of the caps that I can afford, I find MIT to be the best sounding to my ears. I built Pass's Bride of Zen preamp, and initially tried Rifa output and input caps. I "upgraded" them to Infinicaps, and the magic died. I immediately put the Rifa's back in. Later I swapped the Rifa's with Multicaps, and was impressed.
When I did Curcio's "Daniel" preamp project, I used MIT's .47uf/400volt as coupling in the phono and line circuitry. Unfortunately, they don't make values small enough to be used in the RIAA section of phono, so I used Wima's. The results were outstanding, especially phono. When I undertook this latest tube amp project, my mind was made up regarding the useage of MIT whereever possible. I am sure there are better caps, but you get into serious money when you talk paper in oil, teflon mit's, etc.
mg16

 

[dice45]

Are those Teflons magnetic?

...I have a friend that swears that the Multicap Teflon caps are sufficient to start a new religion, and though ...

And the cost...holy smokes...the cost...

Grey,

ask your friend whether his Teflon caps have a metallic housing with hermetic feed-throughs for the leads .. those are usually military. If so, ask him whether this metallic housing is attracted by a magnet. If so, stop dreaming of those Teflons.

Besides that, 1µF MKV costs about 10 Euro, not cheap, but no rip-off either.

My test showed caps have to be nonmagnetic. Under all circumstances. Otherwise you may get gorgeous hifi but music suffers and so do you.

BTW, all caps of my test were measured, either by me or by AllenWright. Those sounding best also were measuring best on the bench as far as tangens delta and dielectric absorption were concerned.

I have the same anti-magnetic opinion about resistors but I havn't proven that yet.

I am a capacitor loather and try to get rid of caps where I can. But if I have to use them, I use micas and proQuality polystyrenes for small values and MKVs for values from 0.1µF up to 47µF (already huge!!). I try to avoid electrolytics.

The preamp/poweramp combo I am planning at the moment has one coupling cap between MC cartridge and speaker (altleast for the fullrange speaker).

Greets,

 

[Thatch_Ear]

Congrats on your 45 amp! Surprising how much bass you can get out of a couple of little watts. As far as small value coupling caps I have had good luck with WIMAs though they are a pain because of the circuit board length leads. I don't think I paid much for them and it took a long time to get around to them but so far I like them untill I like something better. If you are happy with your caps see if you can get some of the WWII 6SN7s labeled VT231s. They are supposed to be the best, and are lots easier to install than new caps.

 

[mg16]

45 diy tube amp

Thanks!
I was really surprised that the bass was so good. Both in terms of low frequency extension, and especially impact. When the drummer really laid into his floor tom on one disc, you felt the low pressure wave hit you, and It sounded very real and close to live sound. I play a set of Yamaha acoustic drums on weekends with a 6 pc. working band, and am familiar with how drums sound live.
Having 104 db/1w Lascalas helps with that 2 watts also.
I expected a great midrange, and got it, but bass was a nice surprise.
I thought most tube amps rolled off the top end, but the highs seem to go well beyond my hearing ability.
I built the amp in 3 sections, testing each section before continuing the build process.
I first built and tested the main power supply,B+, and 3 filiment supplys . Measured voltages with my Wavetek meter. All the voltages,(except the rectifier filiment), were on the high side, which I attributed to the fact that the power supplys were unloaded .
After I wired and mounted the 6sn7 driver circuitry, I powered the amp up, and fed a 1vpp triangular wave input signal to the input jacks, and looked at the output with an oscilloscope.
I was going to measure the -3db rolloff points. I couldn't find one at the low end. It was flat with a voltage gain of 10.6 from a few hertz, out to over 150,000 Hz. The triangular wave stayed very clean across the band.
6SN7 is a great ,wideband, amplifying tube.
I then hooked up the opt's, and plugged in the 45's, but didn't measure the final output at the speaker terminals, as I was too anxious to listen to it.
Later I'll put a 20 watt, 8 ohm, non-inductive wire wound resistor across the output terminals, and take some measurements using the scope, with square wave, triangular wave, and sine wave inputs.
Right now, I don't want to pull the amp out of my system.
Thanks for the tip on the vt231. I will definately try some tube swapping down the road.
mg16