• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

6V6 Marblewood after a 16y break

PabloL

Active Member
What? My last post was 16 years, 1 masters degree and 2 kids ago?!? Well, I made the mistake of picking up a soldering iron, smelled the fumes, and now I have some Edcor transformers and a giant box of parts on the way to my home. I looked at a couple of PP schematics and of course couldn't find one I liked, they all seem overly complicated, or full of parts that when I look them up are EOL, or are mono-blocks... Then I came across Matt's (Suncalc on DIYaudioprojects) 6V6 SEUL amps and settled on the Marblewood. I followed the BOM, except I went with the larger GXSE15-6-5K output transformers, my speakers are 6ohm, and for some weird reason I could find 5.1k or 4.99k resistors but not 5k, but I'm trying not to be OCD about it. Ordering from Edcor had some new options, there was the option to VPI Varnish the power transformer, a XPWR106 for an extra $20, seemed like a good idea and to buy another set of endbells to mount the output transformers vertically, which I got to save space. As for tubes, his design allows for a lot of options so I went with 6W6s and a 12BH7 which someone reported good stuff about. I'll likely roll more tube combos very quickly. No pictures yet, so here are the schematics and the BOM.
 

Attachments

  • MW_Schematic_ps.jpg
    MW_Schematic_ps.jpg
    150.1 KB · Views: 90
  • MW_Schematic_amp.jpg
    MW_Schematic_amp.jpg
    86.4 KB · Views: 91
  • 6V6-Marblewood-Parts-List.pdf
    6V6-Marblewood-Parts-List.pdf
    62.2 KB · Views: 20
Register to hide this ad
Even with the UL connection on the output tube it seems like you should still have some negative feedback. I've heard a couple of UL amps without feedback and thought they had a sharp and wiry sound. You would lose some power but I suspect it would sound better with the tubes triode wired.

If you are considering the 6W6 be aware that the max screen voltage is 200v per the data sheet. It will see close to the full B+ if UL connected.
 
These are nifty little amps and sound great. Below is a link to my build that may be of some use. One particularly note: as built (per Matt's schematic), there was a good deal of low end roll off (down 7dB at 20 Hz) and high distortion, especially on the low end. Upon discussion with the group, I added some "Schade" type feedback and greatly improved frequency response and distortion levels for the amp. Happy building!
Dave
PS I had a few Jan 6V6's around, so I went with those and the 12AU7 option.

 
I read through your thread Dave and the moral of the story was, a 300k "Schade" feedback resistor, removing the bypass cap, and regaining lost gain with the 12AX7. Sounds interesting, thanks for the tip. I realized today that the 350ohm 5w Rd wasn't on the BOM and I forgot to order it, I can get some more tubes and a couple more resistors while I'm at it. I'll solder my bypass caps so I can remove them without much trouble and do some A/B testing, unfortunately I don't have any test gear anymore so It'll have to be my ears unless there has been some easy new testing solutions in the 16years, I don't feel like buying a used scope. With regards to the 6W6 operating out of specs, honestly, the maker of the amp claimed that it was designed with the idea being able to use many different octal tubes in the family, and people had mentioned getting good results, and I hadn't gotten around to doing any math yet... I'm a historian, I super hate math... I'll drop the Matt a note about the 6W6 screen voltage and maybe pick up some 6V6s while I'm shopping. Yay, I forgot about the ever evolving nature of tube projects.
 
Last edited:
Yeah the only thing I would change is try a 0.22uf coupling cap (I really like the Mundorf EVO aluminum oil). Although the 0.47 isn't gonna hurt, just bigger than it needs to be IMHO. And Dave has a good point, start with 300K of schade feedback (I use 3W vishay PRO) and adjust up or down 30K to taste. 330K will be a little brighter, 270 a little less bright. And doing this might require a different driver tube, I really like the 5751 in applications like this.

And great choice on OT, those 15W Edcor 5K SE are GREAT sounding transformers, I use them in my EL34 builds. Finally I've also found pentode tubes used in amps like this sound and perform best when they are running around 80-85% of the max rating. Under driving a "bigger" tube in an SE amp makes for more distortion and just a weak sound IMHO.
 
Instead of a 12ax7, I might go with something like a 12AY7 if you do actually need more gain here. Its about 50% more gain than a 12AU7, has good linearity, and can swing a fair bit more current than a 12AX7 so the feedback resistor will not be much of a challenge for it to overcome. Never fooled with local feedback though so take that as an armchair quarterback opinion.

A 100k plate resistor seems like an overly high value for a 12AU7 or other similar-ish tube anyway. I wonder if they were trying to squeeze more gain out of it.

I wouldn't worry too much about the screen ratings on a 6W6 though. Its got triode conditions with plate and G2 tied together at 300 volts. UL will be similar enough that if its good for triode mode it should be fine for UL.

other octals you might consider are 6K6 and 6F6. They all bias a little different so the cathode resistor probably needs changing to get any of these really optimal but both of those are semi-forgotten types that go for small money and work at similar enough loading and voltages to a 6V6.
 
The 6aq5/el90 is a slightly downrated 6v6. There are bunches of great tubes out there for very little money. I'm also a fan of the cv4043/6bw6 from Brimar. They seem to be a drop in replacement for a 6v6. I run them in ultra linear push pull 300v 40ma and I love the sound. Also super cheap.
 
All the parts for a couple different configurations are ordered. I figure I have enough keep busy for a while once I get everything in the mail. Are Edcor's lead times still pretty accurate? About a month or so? I hate the wait. I'll start off with the stock build, check it out, make sure its running right, listen for a bit and then add the plate to plate feedback and see if my highly trained and accurate vintage ears can hear the difference. It'll be fun, thanks for all the input, standby for pictures, it'll likely be industrial.
 
Looks like a fun project! As mentioned above, it will need feedback if hi-fi performance is the goal. I would also urge you to replace the 12AU7 with a more linear tube. Types like the 6CG7, 12BH7 and 6SN7 are better choices.

Jack

12AU7-6SN7-comp.jpg
 
I have the a 12BH7,12AY7, a pair of 6W6s and 6V6s to try some different combos of tubes. I didn't even pick up a 12AU7. I also got a couple different feedback resistor values around 300k, I figured this would be a good little amp to use to learn how to use my Oscilloscope and AWG (probably a Siglent SDS804x HD and a SDG1032X, which should be all I ever need) Ultimately I want to know how to use my test gear better before building a higher wattage PP amp.
 
Well, it's wired up. It probably could be better, but as a first project after a long break I'm pretty confident I got everything in the right spot. The last thing to do is to put in the rectifier tube, power on and check the voltages. The only real changes I made were, removing the caps on the input stage, adding the discussed plate to plate feedback of 330k, and I grounded the output commons.
 

Attachments

  • 1000011768.jpg
    1000011768.jpg
    160.8 KB · Views: 33
  • 1000011765.jpg
    1000011765.jpg
    105.3 KB · Views: 31
  • 1000011766.jpg
    1000011766.jpg
    112.7 KB · Views: 31
That plate to plate feedback will help tremendously. I've built a couple of SE like that and they sound really nice. The issue with it is that it doesn't include the output transformer but if you have a good one it works great.
 
Well something is up, I was measuring the voltages and things seem good up to the left channel output transformer. It was much higher at the red wire where the B+ goes in (290v Left vs 275v Right) but the blue wire to the plate was way low (67v Left vs 265v Right) I looked at the tubes and noticed the right tube was red plating with the 6w6's in so I turned everything off. I decided to check the resistances around the output tubes and noticed If I measured between the Red and Blue wire on the primary side of the output transformer, the resistance was much higher on the left output transformer 432 kohm (Left) vs 140 ohm (Right,) additionally on the left there is no reading when measuring between the Red wire and the screen tap (white/Blue,) its 60 ohms on the right side.
 
Sorry to say, but it sounds like you've lost an output transformer. It also sounds like something in the circuit might have done it. The left side OPT is open, and the right side output tube is red. I'm guessing the left side output tube might have been red too before the transformer opened.

Jack
 
Back
Top Bottom