Any budding PCB makers ?

Dippy

Moderator
Well done Old Fart - that's a good job and useful info for others.
I think I'd add a proper CC PSU though (more dosh sadly).

Can you tell us a bit more about the UVLED choice?
I've never heard of "straw hat". Is this some PIXAXE Forum (or Ebay) term?
Do you mean 5mm or T-1 3/4 or someting else?
Power? Viewing angle?

It would be nice to know as it's hard to compare when Ebay vendors use 'home-made' terminology.
And another worry is whether the bucket-load of cheap LEDs they get this week is the same as next week.

And why have you got pictures of ladies in the background in post#37 photos?
Old Fart or Old Devil? ;)
 

the old fart

Senior Member
Well done Old Fart - that's a good job and useful info for others.
I think I'd add a proper CC PSU though (more dosh sadly).

Can you tell us a bit more about the UVLED choice?
I've never heard of "straw hat". Is this some PIXAXE Forum (or Ebay) term?
Do you mean 5mm or T-1 3/4 or someting else?
Power? Viewing angle?

It would be nice to know as it's hard to compare when Ebay vendors use 'home-made' terminology.
And another worry is whether the bucket-load of cheap LEDs they get this week is the same as next week.

And why have you got pictures of ladies in the background in post#37 photos?
Old Fart or Old Devil? ;)
Hi Dippy,

The power supply I used was one I had kicking around the garage. 24vdc @ 1.2amp, was removed from industrial machine being scrapped, about 20 years ago. alternatives on ebay for about £15.

UV's were chosen because of the output, 3.2-3.8v, 20ma, 395nm, 3000mcd.and the price.
top hat is the shape of the led, but I used standard shaped 5mm ones.

Viewing angle, not a clue...experimented with blank sheet of paper on top of plexiglass, then moved led pcb up and down underneath until a uniform pattern showed on paper. ended up at around 6 inches.

Pics are from old Retro magazines, not rude, but nice...http://pdf-giant.com/... make sure you have adblocker installed.

TOF
 

ZOR

Senior Member
Just one last thing, your PCB looked very well made, just off the record did you use ferric oxide to etch, and what did you develop with. Thanks
 

the old fart

Senior Member
Just one last thing, your PCB looked very well made, just off the record did you use ferric oxide to etch, and what did you develop with. Thanks
'Ferric Chloride Solution PCB Etching 500 ml Working Strength' from Ebay

'Universal PCB Developer' AP01B from Maplin. 10mg per 100ml

TOF
 

ZOR

Senior Member
Many thanks again, as Dippy says this project will be of interest to a lot of others like me. Keep it up
 

the old fart

Senior Member
Timer program.
Code:
;#picaxe 18x

;Timer for UV box

;10 seconds to 9mins 54 seconds

start:

main:

debug

high c.0
if pin0=1 then let bit0=1:else:let bit0=0:endif
if pin1=1 then let bit1=1:else:let bit1=0:endif
if pin2=1 then let bit2=1:else:let bit2=0:endif
if pin6=1 then let bit3=1:else:let bit3=0:endif
low c.0
high c.1
if pin0=1 then let bit8=1:else:let bit8=0:endif
if pin1=1 then let bit9=1:else:let bit9=0:endif
if pin2=1 then let bit10=1:else:let bit10=0:endif
if pin6=1 then let bit11=1:else:let bit11=0:endif
low c.1
b2=b1*10
b2=b2+b0
w2=b1*60
w3=b0*6 +w2

if pin7=1 then
	w4=w3
endif

if w4>0 then 
	high c.6
	high c.7 ; relay
	dec w4
	pause 630 ;tuned for 1 second cycle time
else
	low c.7
endif

low c.6 ; indicate led
goto start




 
Last edited:

ZOR

Senior Member
Thanks for adding the timer section. Like the idea of using thumbwheels, you seem to have found those with bezel fronts, always a problem to hide hole cutouts.
 

the old fart

Senior Member
Thanks for adding the timer section. Like the idea of using thumbwheels, you seem to have found those with bezel fronts, always a problem to hide hole cutouts.
They had been kicking around in the junk box, 'garage', for twenty odd years.

Found a use for them eventually...
 

the old fart

Senior Member
fitted the 12vdc UV boards,

cut a piece of 4mm thick window glass and replaced the plexiglass.

exposed test piece for 3,4,5 & 6 mins.

4 mins gave best result.
 

ZOR

Senior Member
Just having a coffee after sorting out my 7 segment display problems. You look like you have buttoned it up!

I am trailing behind you.

Is your next project going to be about tinning boards, or will you wife have got you by the scruff of the neck? Regards
 

the old fart

Senior Member
I leave the photo resist etch on the tracks.
Supposed to act as a flux when soldering.
I still use 60/40 solder.

finished boards are sprayed with PCB lacquer from Maplins/

Wife is still 2 projects behind, she hasn't seen the latest projects yet..
 

afb

Member
I'm putting out my prototype and low volume production PCBs to be made by the "isolation milling" technique, prior to ordering etched boards in bulk from China. The supplier I found (I know I musn't advertise in this forum) can make boards from your appropriate Gerber files or (at additional cost) scan a layout of the type typically published in an electronics magazine. He can do double sided, but of course the holes are not plated through, but if you make sure your vias are only at (accessible) component legs then you can solder the PCB on both sides. Again at additional cost you can have all the unused copper lands removed to simulate an etched board or if you need more isolation than 0.3mm for high voltage projects.

I've attached pictures of a panel of small PCBs made by the isolation milling method, plus a larger board where I had all the excess copper milled away.

P1070245s.jpgP1070241s.jpg
 

nick12ab

Senior Member
I'm putting out my prototype and low volume production PCBs to be made by the "isolation milling" technique, prior to ordering etched boards in bulk from China. The supplier I found (I know I musn't advertise in this forum) can make boards from your appropriate Gerber files or (at additional cost) scan a layout of the type typically published in an electronics magazine. He can do double sided, but of course the holes are not plated through, but if you make sure your vias are only at (accessible) component legs then you can solder the PCB on both sides. Again at additional cost you can have all the unused copper lands removed to simulate an etched board or if you need more isolation than 0.3mm for high voltage projects.

I've attached pictures of a panel of small PCBs made by the isolation milling method, plus a larger board where I had all the excess copper milled away.

View attachment 16175View attachment 16176
I had boards made by the "supplier you found" when he was trialing the service and only charging £5 for a 160x100 board. The only issue I had was that sometimes large areas of copper could be shorted to tracks or even tracks shorted together where spacing of the tracks caused a thin sliver of copper to be left between the tracks and the ends detached from the board, and made contact with the tracks. So it's important to quickly check everything with a multimeter before and after soldering the components.
 

afb

Member
Sound advice Nick - I've had six batches of different design boards processed now without a problem, but I saw the potential for shorts with detached 'slivers' at the outset and I always check the milled areas with a jewellers eyeglass - not a big issue for the small volumes I use. I'm not into high density stuff and I make the tracks as wide as possible so there is relatively little space between adjacent tracks and the two 0.3mm milling traces generally overlap so there are no slivers left behind. And yes, careful and accurate soldering is a must as far from joints being surrounded by solder resist, they are surrounded by solder magnets! Update - his milling is now sold at £1.50 per 10 sq cm (single sided) or £2.50 for double sided, so my prototypes only cost a few quid - it would be nearer to a hundred the etched PTH silk-screened route, though you'd get quite a few PCBs for that money.
 

John West

Senior Member
We have been conducting a series of classes at our Hackerspace club, everything from CAD schematic drawing through micro-controller programming, the complete sequence from start to finish. I helped with the PCB making class yesterday. About a dozen folks learning the "iron on from a laser print" method. All the bds came out functional, although these were just single-sided surface mnt bds without any holes, so they're kinda hard to mess up.
 

AlbertZ

Senior Member
My First Attempts at Making PCB’s using the Positive Photo Resist Method

My First Attempts at Making PCB’s using the Positive Photo Resist Method

I built an exposure box. The box was built from ½” poplar sides and ¼” poplar top and bottom. Inside dimensions of the box are 4.8” x 7.25” x 3” deep. I used 96 UV LEDs arranged in a 12 x 8 matrix mounted on a printed circuit board. The printed circuit board mounts into the floor of the box on ¼” standoffs.
The LEDs had a luminous intensity of 1200 mcd and a viewing angle of 160 degrees. The broad viewing angle was important because it projected a nice even light at an exposure distance of around 3”. The LEDs were wired in a series-parallel arrangement with 3 LEDs in series with a 100 ohm current limiting resistor.

http://www.ebay.com/itm/231040487215?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649

A 4 mm acrylic plastic window was fitted into the top of the box for exposing the artwork. My projects are not very big and this little exposure box will nicely accommodate a 100 x 150 mm board. I was warned about using regular window glass as it attenuates UV wavelength light.
I mounted a small aluminum enclosure on the side of the box for mounting the ON-OFF switch and a jack to plug in the 12 VDC power supply. I use a handheld stopwatch to time the exposure.

I found a source of inexpensive pre-sensitized boards on e-bay which makes this whole venture worthwhile.

http://www.ebay.com/itm/400569497455?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649

DSC00080.JPG
 

AlbertZ

Senior Member
My First Attempts at Making PCB’s using the Positive Photo Resist Method

Step 2:
I made a test strip wherein I exposed a 1” x 6” strip in one inch increments of 30 seconds for each increment. This meant that the first increment was exposed for 3 minutes and the last increment was exposed for only 30 seconds. After developing the strip in a NaOH solution (10 g/liter), only the last 30 second patch was under exposed. This means that anything over 1 minute will be sufficient exposure. Developing time was about 30 seconds and this is temperature dependent.

Step 3:
Next I exposed a pre-sensitized board with the artwork for a circuit I had developed. The artwork was printed on regular transparency using a laser printer. The artwork was doubled up because when I held a single transparency up to the light it appeared to be too porous. I exposed the transparency for 2 ½ minutes. The end result was an almost perfect exposure except that some of the UV leaked through the blanked off areas. In the future, exposure will be limited to 2 minutes or less.

Step 4:
Etching the boards using Ferric Chloride. I stumbled on a method where I float the boards on top of the solution. Total etching time is 15 to 20 minutes depending on age of the solution and temperature. I still use the toner transfer method for silkscreen, but I am thoroughly delighted with the photo resist method for the traces.

DSC00082.JPGDSC00083.JPG
 

the old fart

Senior Member
Hi AlbertZ

I used acrylic sheet in my UV box, but found if a bugger to keep absolutely clean and clear.

I changed to a 4mm window glass, which still worked just as well, and can be cleaned with 'spirit clean'.

If you place a blank sheet of paper on the plastic sheet you can see if there are any low UV spots.

I print using an inkjet printer, using Jetstar premium transfer sheets. £15 for 10 A4 sheets. But I cut them down to A6 for most projects.
Set printer to, photo matt, black only, turn off high speed print.

I usually buy my boards from Maplins, £12 for single A4 board.

I'll send off for the ones that you are using and see how I get on with them.

I've not tried silk screen printing....yet.

TOF
 

John West

Senior Member
AlbertZ, one small suggestion. I see you have a few thin traces on your bd layout where they don't really need to be thin. While the process you're using clearly can reliably make very thin traces, it's usually better to use wider traces wherever there's room for them. It's the old "belt and suspenders" theory as applied to PCB layout.
 

AlbertZ

Senior Member
AlbertZ, one small suggestion. I see you have a few thin traces on your bd layout where they don't really need to be thin. While the process you're using clearly can reliably make very thin traces, it's usually better to use wider traces wherever there's room for them. It's the old "belt and suspenders" theory as applied to PCB layout.
Hi John

You're absolutely right, there was only one trace that needed to be thin where I squeezed between two IC pins. The others, well I just wanted to see how fine the photo resist method would reproduce. Needless to see, I was very pleased.
 

AlbertZ

Senior Member
Hi AlbertZ

I used acrylic sheet in my UV box, but found if a bugger to keep absolutely clean and clear.

I changed to a 4mm window glass, which still worked just as well, and can be cleaned with 'spirit clean'.

If you place a blank sheet of paper on the plastic sheet you can see if there are any low UV spots.

I print using an inkjet printer, using Jetstar premium transfer sheets. £15 for 10 A4 sheets. But I cut them down to A6 for most projects.
Set printer to, photo matt, black only, turn off high speed print.

I usually buy my boards from Maplins, £12 for single A4 board.

I'll send off for the ones that you are using and see how I get on with them.

I've not tried silk screen printing....yet.

TOF
HI TOF

That's the beauty of the wide angle LEDs - they provide a nice even light without any high or low spots.

I read somewhere that you can use semi-transparent Vellum sheets. These are cheaper than the transparent plastic sheets. I'm in your category - retired on fixed income so I have to watch how much I spend. Fortunately I have 50 years worth of components accumulated, but its things like PCBs and enclosures that can run into money.
 

John West

Senior Member
Ah, good idea, Albert. I've done that before on bds where I was testing the process, but I just did a series of ever thinner short traces side by side in the unused ground-plane area to see at what trace width my bd making process became unreliable. That way it didn't affect my functional layout and gave me a good idea of what I could possibly get away with if I needed to.
 

AlbertZ

Senior Member
AlbertZ, one small suggestion. I see you have a few thin traces on your bd layout where they don't really need to be thin. While the process you're using clearly can reliably make very thin traces, it's usually better to use wider traces wherever there's room for them. It's the old "belt and suspenders" theory as applied to PCB layout.
Wouldn't you know it. I was trying to troubleshoot this board and it was driving me crazy. The problem turned out to be an extremely tiny break in one of my "thin" traces. Needed a magnifier to see it and verified with a continuity test. Yes indeed, it's "belt & suspenders" wherever and whenever possible.

Al
 

John West

Senior Member
Even when my home-brew bds look good I still "beep" them with an ohmmeter before I populate them. It takes a few minutes, but it's quicker and easier than circuit troubleshooting.
 
Top