MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Buchla 100 system diy build
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2, 3  Next [all]
Author Buchla 100 system diy build
APETECHNOLOGY
hello, i am going to chronicle every aspect of my adventure in building a Buchla 100 system.
i am building from the information found at

http://ebolatone.blogspot.com

i am starting with the etching of the pcbs .
i have never made pcbs, this was the first time.
so i find it interesting...
APETECHNOLOGY


i first printed out the art on transparency film.
i alined the art onto presensitized kinsten pcb material.
i then exposed it to uv for 5 minutes
i then developed the pcb in the kinsten developer bath.
APETECHNOLOGY


this shows the developed art.
APETECHNOLOGY


this shows the etching tank .
i etched in 2 parts hydrogen peroxide to 1 part hydrochloric acid.
APETECHNOLOGY


this shows the etched pcbs .
APETECHNOLOGY


i used small drill bits for drilling, i bought this set from eBay.i was very happy with it.
APETECHNOLOGY


this shows the drilling, i am using my second milling machine "not so nice" for the drilling. the fiberglass can be rather corrosive.
APETECHNOLOGY


this is how i checked to make sure i drilled all the holes.i held it up to a light.
APETECHNOLOGY


the last process after drilling i did was to treat the boards with liquid tin.
super creepy toxic stuff. gloves are highly recommended for all these processes involving liquid chemicals!
APETECHNOLOGY


this was the late boy to the party,
thank you Mike Peake!
ok, now to the nightmare of sourcing all the parts... yikes woah
Peake
Have fun! Note that the 158 and 144 transistor pads are a bit close and it's easy to do solder bridges so check them after building with a bright light behind the board (thanks Tim Stinchcombe for that one!) Your etches came out well, good stuff.

Also note that defluxing with alcohol etc. can begin to remove the liquid tin so use water clean solder if possible, I still have to try that.

Also a slightly larger drill bit (1mm or so) for the 106 and 156 PCBs where the pots mount on the PCB for the two side pins, as well as for all wiring connections, allows a tinned wire to go through the PCB with ease.

Note on the image of the 144 with the light behind it to check for undrilled holes, the lower right corner, the horizontal trace above where the D case 15uF cap goes on the trace for the large output tantalum (+ symbol); you can see a broken trace. This is one of the kickers in debugging; a slight trace break can pass some voltage but not all. I check against a bright light after developing the exposed PCBs for breaks and use a Sharpie to fill in any issues prior to etching. Check again with a bright light behind it after etching. Just jumper such areas with a clipped resistor lead or use the lead of a stuffed component in the path to do so. Check all of your PCBs for breaks.

There is so much that can go wrong in the multiple stages of doing positive resist PCBs. You did really well.
RadioTelefonik
Looking good so far! I'm interested in doing a 100-series build eventually, so you can be sure that I'll keep my eyes on this thread!
Peake
J201 -can- be used to replace 2N4339; note that smallbear currently has them for less than a dollar while Mouser has them for $3.30 each.
Jarno
+1 on checking for trace breaks and shorts, I had a triple low-pass gate that I spent a lot of time debugging on, only to leave it in a bin for a year, and then noticing a tiny spot where the copper between traces wasn't completely etched away.......
I used the layout from Mark Verbos' old website, and it has a few spots where traces run close to the outside edge (which is ground, IIRC).
APETECHNOLOGY
"Note on the image of the 144 with the light behind it to check for undrilled holes, the lower right corner, the horizontal trace above where the D case 15uF cap goes on the trace for the large output tantalum (+ symbol); you can see a broken trace. This is one of the kickers in debugging; a slight trace break can pass some voltage but not all. I check against a bright light after developing the exposed PCBs for breaks and use a Sharpie to fill in any issues prior to etching. Check again with a bright light behind it after etching. Just jumper such areas with a clipped resistor lead or use the lead of a stuffed component in the path to do so. Check all of your PCBs for breaks. "

i dropped the ball Peake, the whole point of that picture was to circle the broken trace! thanks again for all your hard work and guidance on this project.
thumbs up
APETECHNOLOGY


Peake, this is a picture of the 410 quad mixer from Ebolatone,
i am confused about the orientation of the 2n4339?
it would seem that the left hand side should be reversed.
i had to use heat shrink and some mechanical gymnastics to make 2 of them work.
in the photo, both sources go to the right on the top.
also what is the value for the potentiometers ?10k linear?
thanks !
Eric the Red
Wow - self etching PCBs looks like a chemically nasty process. This is truly an awesome build, keep up the good work.
Peake
Pots are 10K linear. The FET's Gate goes to R6, a 10K resistor. G is the unlabel'd pin in the center facing to the inside of the PCB. G is the one which matters. Note that J201 can usually be DSG pinout. If you know the manufacturer of your FET you can find the datasheet. If not, the Idss check can reveal it.

https://rubidium.dyndns.org/~magnus/synths/companies/buchla/Buchla_410 0_8_200.jpg
Jarno
Eric the Red wrote:
Wow - self etching PCBs looks like a chemically nasty process. This is truly an awesome build, keep up the good work.


Haven't tried tin plating myself, but etching really isn't that bad, I used toner transfer rather than silkscreening though.

Drilling is a little bit bad, that dust is something you need to not breathe in.
Laughing
I've used toner transfer as well, but the print quality was really shit, unfortunately. I know the darkroom developer method exists as well, but what else besides that? Doesn't the darkroom method depend on a good print, too?
APETECHNOLOGY
Laughing wrote:
I've used toner transfer as well, but the print quality was really shit, unfortunately. I know the darkroom developer method exists as well, but what else besides that? Doesn't the darkroom method depend on a good print, too?


for the darkroom method i just printed out a transparency on a laser printer.
turned out the lights, put on a head lamp that had a red led.
opened the presensitized board.
alined the transparency on the glass with the pcb,and exposed it to uv light for 5 minutes.
seems long, i don't really know the wavelength of a bug zapper fluorescent lamp! seems it would take the same amount of time with a regular lamp.
i am really happy i did the extra work.
i did toner transfer a couple of years ago, i was not happy with the process or the end result!
Ayab
Lovely result with the etching. You can easily add a parts placement silkscreen as follows - If your boards are not stuffed and you have a good parts placement this may save you some time soldering and helps avoid mistakes plus can help with troubleshooting.

https://www.youtube.com/watch?v=M_m1YBTzgvY

(n.b. you may want to turn the sound off - sorry not my video)

It works really well - colour or black and white. You can just rub it off with acetone and try again if it doesn't transfer properly. I have had good results with much better detail than I expected.

What I do:

1. Load pcb traces layout into photoshop then create a new layer with the parts placement layout and 'free transform' the parts layer to match it to the traces perfectly. (It may be necessary to reduce the opacity of one of the layers during this process.)

2. Sometimes I do a bit of tinkering with the parts image to make the print out of the components on the silkscreen stand out more. Sometimes inverting the colours makes a load of difference. If you are interested pm me the placement and I can let you know.

3. Mirror/flip horizontal 180 the parts placement layer and print out (without the traces layer, obviously) onto normal laser paper.

4. Clean pcb carefully with a little washing up liquid to degrease finger marks and then only touch edges.

5. Holding the etched & drilled pcb up to a bright light match the position of the parts layout exactly (print face down) and tape with a couple of bits of masking tape.

6. Wipe plenty of acetone/nail varnish remover over until it is soaked and pressing and rubbing a little on the detail with your fingers - the trick is not to break up or move the paper. It is not difficult to see just when the surface of the paper is starting to break up into bobbles so stop at this point or before.

7. Leave to dry - then its done it should peel right off v.easily
APETECHNOLOGY
Peake wrote:
Pots are 10K linear. The FET's Gate goes to R6, a 10K resistor. G is the unlabel'd pin in the center facing to the inside of the PCB. G is the one which matters. Note that J201 can usually be DSG pinout. If you know the manufacturer of your FET you can find the datasheet. If not, the Idss check can reveal it.

https://rubidium.dyndns.org/~magnus/synths/companies/buchla/Buchla_410 0_8_200.jpg

thank you for your answer, i used odd 2n4339 for the fets.
i am sourcing parts for the 158,
i have some questions,
22pf ish ,this was a variable capacitor. originally 35pf? its hard to tell from schematic.what do you use for this part?
.047uf= film cap?
47 pf= ceramic?
and last question...
out of 15- 2n4339, i have 4 that are around idss .57ma will the trimmer be able to compensate for this?
the j201 i have "from small bear" have a idss of .90 or more!
thanks!
APETECHNOLOGY
Ayab wrote:
Lovely result with the etching. You can easily add a parts placement silkscreen as follows - If your boards are not stuffed and you have a good parts placement this may save you some time soldering and helps avoid mistakes plus can help with troubleshooting.

https://www.youtube.com/watch?v=M_m1YBTzgvY

(n.b. you may want to turn the sound off - sorry not my video)

It works really well - colour or black and white. You can just rub it off with acetone and try again if it doesn't transfer properly. I have had good results with much better detail than I expected.

What I do:

1. Load pcb traces layout into photoshop then create a new layer with the parts placement layout and 'free transform' the parts layer to match it to the traces perfectly. (It may be necessary to reduce the opacity of one of the layers during this process.)

2. Sometimes I do a bit of tinkering with the parts image to make the print out of the components on the silkscreen stand out more. Sometimes inverting the colours makes a load of difference. If you are interested pm me the placement and I can let you know.

3. Mirror/flip horizontal 180 the parts placement layer and print out (without the traces layer, obviously) onto normal laser paper.

4. Clean pcb carefully with a little washing up liquid to degrease finger marks and then only touch edges.

5. Holding the etched & drilled pcb up to a bright light match the position of the parts layout exactly (print face down) and tape with a couple of bits of masking tape.

6. Wipe plenty of acetone/nail varnish remover over until it is soaked and pressing and rubbing a little on the detail with your fingers - the trick is not to break up or move the paper. It is not difficult to see just when the surface of the paper is starting to break up into bobbles so stop at this point or before.

7. Leave to dry - then its done it should peel right off v.easily




thanks for the tip!

thumbs up
Peake
APETECHNOLOGY wrote:
Peake wrote:
Pots are 10K linear. The FET's Gate goes to R6, a 10K resistor. G is the unlabel'd pin in the center facing to the inside of the PCB. G is the one which matters. Note that J201 can usually be DSG pinout. If you know the manufacturer of your FET you can find the datasheet. If not, the Idss check can reveal it.

https://rubidium.dyndns.org/~magnus/synths/companies/buchla/Buchla_410 0_8_200.jpg

thank you for your answer, i used odd 2n4339 for the fets.
i am sourcing parts for the 158,
i have some questions,
22pf ish ,this was a variable capacitor. originally 35pf? its hard to tell from schematic.what do you use for this part?
.047uf= film cap?
47 pf= ceramic?
and last question...
out of 15- 2n4339, i have 4 that are around idss .57ma will the trimmer be able to compensate for this?
the j201 i have "from small bear" have a idss of .90 or more!
thanks!


-Variable capacitor in 158: didn't seem to do anything so I went for a median value.
-.047uF = film cap.
-47pF = ceramic cap.
-Idss of the E101 in the 158's sine shaper was supposed to be 0.5 so try the 4339.
-Pre55ure pointed out that the 470r to ground just after the sine shaper can be made a trimmer to zero in. Going off of memory and not the schematic.
-Your J201 are at the extreme high end of possible Idss, keep them for later builds with a sine shaper (285, 208, 258, 259).
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2, 3  Next [all]
Page 1 of 3
Powered by phpBB © phpBB Group