BA-6A time constant board design

[Martan]

I’ve received the time constant boards from oshpark and have one populated. I missed one connection but it was easy to jump. I’m thinking I won’t be able to evaluate it tomorrow Neil I get the whole thing working. I just got some new tubes to try out.

 

[scott gradel]

I’m building this as well bought in 2019 and have zero documentation or even know if I have all the boards or if it was supposed to just be one it came with the legacy package he sold at the time and was to good of a price to pass up so I been just filling the easy stuff hoping I can find info

 

[scott gradel]

I’ve received the time constant boards from oshpark and have one populated. I missed one connection but it was easy to jump. I’m thinking I won’t be able to evaluate it tomorrow Neil I get the whole thing working. I just got some new tubes to try out.

And I’m super interested in if that board actually worked out for you

 

[JMan]

I’m building this as well bought in 2019 and have zero documentation or even know if I have all the boards or if it was supposed to just be one it came with the legacy package he sold at the time and was to good of a price to pass up so I been just filling the easy stuff hoping I can find info

It's just the one big board. The "Fairchild time constants" (which are referenced in the BOM) were implemented on a separate board that he stopped selling, and that's what Martan was recreating with his board.

If you send me a PM to remind me, I can get you the build docs.

 

[Martan]

HI,

Here's a link to the Dropbox folder I have on the Drip BA-6A. Should be helpful.

https://www.dropbox.com/sh/5zqwlmwtoxqxatx/AADJ3mRsrDth5UTy6Klj_u8ua?dl=0

 

[nielsk]

This was all in DaveP's thread on his build (if the image links still work, if not he would likey send it to you), so easy to build out on a switch.
No need for a PCB

 

[Martan]

Here is a board I drew up in Eagle for the Time Constants for the BA-6A. It is based on me having the CollectiveCases front panel for the build that has more markings for the Time Constants than the original unit. There are enough tick marks on the front panel to add the Time Constants from the Fairchild 660, but its hard to find people who have tried/recommend this. The parts fo the Fairchild constants are listed in the Drip BA-6A build guide. Either way, I wanted to try my hand at laying out a board and have it fabbed (OSHPark). Beware that some traces were missed on the OSHPark run (my fault) and need jumpers. I have tried to correct them on the attached pcb.

As nielsk points out, the original 3 settings for the BA-6A would be pretty easy to build onto the back of a switch. I have flipped through DaveP's thread, but can't see any of the images. My pcb is for a GH-7107 switch I got from digikey. The 660 Time Constants required a more complex switch.

All the best,

Marty

 

[xazrules]

Hi Martan! I’ve build both the 660 and ba6a, and i’d like to suggest you to not mix up the sidechains because the one in the 660 is very different from the ba6a/436/175 topology. The TC is very dependent on what is driving it. with well matched 6sk7 you can push it very fast but you will never reach position 1 of a 660.
The 100nf to ground is your attack time that depend on how fast the 6v6s charge it. you can experiment by lowering (faster)or upping (slower)this cap, or just place a resistor in series after the 6al5. Your release time is the 3.3meg to ground but Adding a pot or some resistors on a switch you can make it variable and go by ear!
Love to hear when someone fjnish a ba6a because i almost prefer it to a 660.

Cheers

 

[JMan]

I'm not too far off of finishing mine now, but summer is an extremely busy time for me at work and so I haven't been giving the build much attention. I'll probably have at least one or two more questions before I get to the finish line, so I'll be checking back in about it in the next couple weeks. Any help will be greatly appreciated!

 

[Martan]

Thanks, xarules! Once I get mine running correctly I would love to experiment with the sidechain values. Something is wrong with the preamp section. It passes sound (weak) but the input control does nothing. I keep hitting problems with this build (several years on and off) and am starting to get the urge to take another crack at it

 

[xazrules]

i'm pretty sure there is an error in the wiring of the dual input pot or an error in the wiring of the input transformer, i'm sorry but i've never seen the drip pcb.

 

[nielsk]

DavePs version has 6 positions, one of which is variable.
It seems the image links are often down in slightly older threads, but as I mentioned it may be worthwhile to reach out to him as he did some clever work and the schematics are clearly drawn and easy to understand.
There is value in learning PCB layout and design and making some to find out what works, but it is also useful the see when it is so much quicker and easier to just build out something very simple on a switch, it works great, saves time & $ and it is they way pretty much everybody who made these things originally did it.

 

[xazrules]

I hope i can share it, this is the DaveP sidechain

 

[Martan]

Thanks, xazrules and nielsk. That schematic will be very helpful.

 

[JMan]

Question - I saw in this old thread that @emrr said a T attenuator needs to be terminated. Since that's what I've got on the output here, is termination advisable?

Also, sorry to be asking questions here that aren't about the time constant board. It's just that so few folks are working on this project anymore, plus there's a certain aversion on the forum to supporting these projects (which I think is justified, but kind of sucks when you're the builder who needs help).

 

[xazrules]

can you post a schematic? you need an output pad after the transformer of the ba6a?

 

[JMan]

Sure, here you go. The question isn't so much about a fixed output pad but about whether the variable attenuator after the transformer, which is a T attenuator, needs to be terminated with 600 ohms.

The question of inserting an optional fixed pad on the output, which isn't in the schematic but is indicated in the build, would be a separate one. I sort of agree with the idea that it seems odd to have a fixed pad after the transformer when there's already the variable attenuator there, but maybe it just puts out a huge signal that needs a lot of taming? I haven't finished the project just yet so that's only a guess.

 

[xazrules]

all right, if i understand correctly you are not sure about the resistor strapped across the output transformer terminals, i think if you have already a T pad you don't need it, but in every schematic (in the sta level too) there is one, usually i shunt the ouput trafo with a larger resistor than 600 ohm, 1k to 3k then i add the variable attenuator, in doing so you are loading just a little bit the tubes.
in the case of the ba6a i've created my own variable attenuator at 200 ohm.

cheers

 

[lain]

I was pulling out my Drip BA6A to finish building it, and I remembered seeing this thread, and thought I should reference it. Then I found one of the time constant boards in the box, along with a DC heater board.

 

[Rob Flinn]

I have been watching this thread from the sidelines, because I'm 80% through a p2p Ba6a project.

One thing I would suggest is that if you need a pcb for the time constant board that you don't rely on just the grayhill switch to support it. I built a Drip 670 a couple of years back & that was one of the aspects of it that I thought was poorly designed. It feels weak & because the pcb is not secured by anything other than the solder connections of the one grayhill, to me it is not strong enough. If it were me I would use something like a bit of angle aluminium bolted to the pcb & secured under the grayhill mounting nut to give it added strength. However, if you're happy with this don't let me stop you.