STA-Level: 6386 to 6BA6 T-Bar wiring

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

morls

Well-known member
Joined
Jul 15, 2013
Messages
257
Location
Australia
Hi,

I'm going with 2 x 6BA6 tubes instead of 1 x 6386 for this build, using a (modified) T-Bar board from Manley Labs. Another user here, FarisElek, uploaded a schematic from Bowie:

6386 adapter wiring.png

As this shows, pins 2 and 7, and 5 and 6 are connected for both 6BA6 tubes. I'm thinking of joining them beneath the board with an uninsulated copper wire, directly from the socket pins. Is this the right way to join pins? Because 5 and 6 are adjacent it's an easy join, but joining 2 and 7 requires the wire to cross the socket. Would this cause any issues?

I imagine the connection would be from 6386 socket 8 to 6BA6 pin 2, and then joining 6BA6 pin 2 to 7 with copper wire. Is this the best way, or do I need the connection from 6386 pin 8 to be equidistant from 6BA6 ins 2 and 7? By this I mean just like the circuit above shows the connections between heater pins 3 and 4 of both 6BA6s and pind 1 and 9 of the 6386.



The T-Bar has 2 PCBs. Here is the lower board, which has the 9-pin extender for the original 6386 socket. I've removed all the copper traces, as this version was wired for the 5670 tube, an alternative to the 6386 that Manley was using at the time this version of the T-Bar was in use. My plan is to connect each pin from the 6386 socket to the corresponding terminal via hookup wire:

T-Bar lower 1.jpgT-Bar lower.jpg



Here is the upper board. The traces on this PCB match the 6BA6 pinout. I've removed one trace, which was between pin 3 of 6BA6 #1 and pin 4 of 6BA6 #2, the heater pins:

T-Bar upper.jpg T-Bar upper 2.jpg



And here are the two boards as they'll be assembled:

T-Bar both.jpg T-Bar both 1.jpg



The heaters for all tubes in this build run at 6.3VDC. Pins 1 and 9 on the lower board are accessible with the top board in place, and I'm thinking about the best way to connect these to the upper board. Because of the layout on the top board, if I join pins 3 on both sockets AND pin 4 on both it will be a criss-cross. I could do one join on top of upper PCB and one bottom of upper PCB. Still a criss-cross but separated by the PCB. I'll also add the 68 ohm resistors on the bottom board, between socket pins and terminals.

Does this all look reasonable? I don't know enough about electromagnetic interference to tell whether I'd be introducing a whole lot of noise if I wire things this way...

Cheers
Stephen
 
Wouldn't it be better to use some EF83's which are cheap, or another popular vari-mu tube, such as the 6BC8, instead of a standard pentode?
 
I'm not familiar with those tubes. From what I've read the 6BA6 works very well in this circuit, and I've got all the tubes ready to go.
 
Well, if you already have everything, why not.

As far as I can tell, the main reason to use a 6BA6 is if you are planning on manufacturing them, therefore you need a steady supply, because they are very common tubes.

They are not really vari-mu tubes, so for a one-off you might find more appropriate tubes for the job.
 
Sounds interesting. I’ve got another T-bar set, so once I get this project up and running I might look at other options.
 
Well, if you already have everything, why not.

As far as I can tell, the main reason to use a 6BA6 is if you are planning on manufacturing them, therefore you need a steady supply, because they are very common tubes.

They are not really vari-mu tubes, so for a one-off you might find more appropriate tubes for the job.
It’s quite well documented that they act exactly the same.
 
If you're replacing what was a 6386 in a circuit then, yep, 6BA6 in triode mode is a good sub.

If your heaters are clean DC then I don't see an issue with what you're doing introducing any extra noise. Manley Labs. do it that way and their customers seem happy enough so...

And 6BA6 pairs work in Fairchild builds in lieu of 6386's so why not in a STA-Level.


Buy a few 6BA6's and then you can get a reasonably matched pair. A curve tracer of some sort would be the high tech way to match but, common sense and some 'trial & error' testing in situ under various states of compression should get you where you need to be.
 
As this shows, pins 2 and 7, and 5 and 6 are connected for both 6BA6 tubes. I'm thinking of joining them beneath the board with an uninsulated copper wire, directly from the socket pins. Is this the right way to join pins? Because 5 and 6 are adjacent it's an easy join, but joining 2 and 7 requires the wire to cross the socket. Would this cause any issues?

I don't think it'll cause any issues. Looks to me like heater pin 3 on the 6BA6 would be the earthy connection when DC fed but check if the datasheet says differently.

Just for the sake of being retentive, when you connect pin 2 (cathode) and pin 7 (suppressor grid), have the actual wire that connects to the lower pcb come from the pin 2 side of the join.

Same when you join 5 & 6 - have the wire to the lower pcb come from pin 5 side of the connection.



I imagine the connection would be from 6386 socket 8 to 6BA6 pin 2, and then joining 6BA6 pin 2 to 7 with copper wire. Is this the best way, or do I need the connection from 6386 pin 8 to be equidistant from 6BA6 ins 2 and 7? By this I mean just like the circuit above shows the connections between heater pins 3 and 4 of both 6BA6s and pind 1 and 9 of the 6386.

See above.


The heaters for all tubes in this build run at 6.3VDC. Pins 1 and 9 on the lower board are accessible with the top board in place, and I'm thinking about the best way to connect these to the upper board. Because of the layout on the top board, if I join pins 3 on both sockets AND pin 4 on both it will be a criss-cross. I could do one join on top of upper PCB and one bottom of upper PCB. Still a criss-cross but separated by the PCB. I'll also add the 68 ohm resistors on the bottom board, between socket pins and terminals.

Does this all look reasonable? I don't know enough about electromagnetic interference to tell whether I'd be introducing a whole lot of noise if I wire things this way...

Cheers
Stephen

Criss-crossing the heater wires is OK :) 👍
 
Last edited:
unfortunately, the threads which have curve traces are too old, and the pics have disappeared.
 
This page describes the approach taken by Manley Labs. I've gotta say, I was blown away by the response to my initial enquiry about getting some boards. I got a very quick reply from the company's president, EveAnna Manley, who put me in touch with the parts department and we went from there. Amazing support for a humble DIYer!

Manley Labs T-Bar
 
If your heaters are clean DC then I don't see an issue with what you're doing introducing any extra noise. Manley Labs. do it that way and their customers seem happy enough so...
Thanks.

Buy a few 6BA6's and then you can get a reasonably matched pair. A curve tracer of some sort would be the high tech way to match but, common sense and some 'trial & error' testing in situ under various states of compression should get you where you need to be.
Christian Whitmore at proaudiotubes has helped me out here. I've got a few spares so I can swap out any noisy ones.
 
My final question about this mod has to do with balancing the two 6BA6s. The board I'm using is the Drip Dual-STA, and there is a balancing resistor wired in to balance the two halves of the 6386, marked below as 100R Balance:

trimmer resistor schematic.jpg

It would be great to be able to have this for the 6BA6s as well. Is there any reason why this would not work?

Cheers
Stephen
 
Roughly 10 years ago I built a prototype sta level using a pair of 6ba6 with a view to doing a small production run having been asked by some clients for them. It worked quite well & sounded good. Unfortunately just as I got the prototype working a recession hit so the project never went ahead.
 
My final question about this mod has to do with balancing the two 6BA6s. The board I'm using is the Drip Dual-STA, and there is a balancing resistor wired in to balance the two halves of the 6386, marked below as 100R Balance:

View attachment 82101

It would be great to be able to have this for the 6BA6s as well. Is there any reason why this would not work?

Cheers
Stephen
If you find a good match for the 6ba6 pair you can normally null any slight thumping with the balance trim on the 6V6's
 

Latest posts

Back
Top