Favourite parts with off-label uses?

[Lasse Langwadt Christense]

lørdag den 11. april 2020 kl. 08.32.10 UTC+2 skrev Ricky C:

Aren't there pretty inexpensive ways of getting prototype quantities of PWBs made? I recall several outfits that panelize multiple designs into one batch so you can order a small number of boards at not a high price.

places like jlcpcb etc. will do 5x 100*100mm dual sided, with silk,mask etc..
so vastly better than anything you can make at home, for $5 the shipping might
be $10 these days

https://pcbshopper.com/

 

[mpm]

On Saturday, April 11, 2020 at 12:22:46 PM UTC-4, Ricky C wrote:

On Saturday, April 11, 2020 at 11:19:17 AM UTC-4, Lasse Langwadt Christensen wrote:
lørdag den 11. april 2020 kl. 08.32.10 UTC+2 skrev Ricky C:

Aren't there pretty inexpensive ways of getting prototype quantities of PWBs made? I recall several outfits that panelize multiple designs into one batch so you can order a small number of boards at not a high price.


places like jlcpcb etc. will do 5x 100*100mm dual sided, with silk,mask etc.
so vastly better than anything you can make at home, for $5 the shipping might
be $10 these days

https://pcbshopper.com/

Someone is asking me to layout a board that is a bit larger than my usual size range and I was concerned the board cost might be getting up there so I checked with these guys. Not bad at all. But while exploring options I noticed some oddities.

When ENIG is selected, not only do they charge $16.00 for 'surface finish', but they add another $9 to the 'board' cost. So it's really $25 extra. Same with 2 oz copper weight. Odd.

Then when I checked the 1 layer price vs. the two layer price they lower the 'board' price just a bit less than $4, then add $4 to the 'engineering fee' so the total is a actually a little bit more!

The price goes up a lot with layers. They don't add much to the "board" fee, but the "engineering fee" goes up a LOT. I guess that's the labor and the board fee is the material.

But the prices are amazing! I had no idea.

Their web page is very clear and seems to cover almost every option. I don't see a way to specify the stackup as in layer thickness. They talk about fiducials being made as holes in the PWB either half way through or all the way through. Wouldn't this just be another drill hole? Or is this made with a higher accuracy?

Checking another board that is pretty small but 6 layer starts to get pretty expensive by the time I add in the smaller hole size and ENIG it's pushing toward $200. Still not bad really.

--

Rick C.

--- Get 1,000 miles of free Supercharging
--- Tesla referral code - https://ts.la/richard11209

Did you look at JLCPCB.com ?

We use them all the time (from China).
Great prices.

We have been buying mostly 2-layer, but we do buy some 4-layer occasionally (but not RF, which we prefer to order domestically)

Typical delivery is 5-8 days, but running more like 8-10 right now days with this Coronavirus thing.

 

[Ricky C]

On Saturday, April 11, 2020 at 11:19:17 AM UTC-4, Lasse Langwadt Christensen wrote:

lørdag den 11. april 2020 kl. 08.32.10 UTC+2 skrev Ricky C:

Aren't there pretty inexpensive ways of getting prototype quantities of PWBs made? I recall several outfits that panelize multiple designs into one batch so you can order a small number of boards at not a high price.


places like jlcpcb etc. will do 5x 100*100mm dual sided, with silk,mask etc.
so vastly better than anything you can make at home, for $5 the shipping might
be $10 these days

https://pcbshopper.com/

Someone is asking me to layout a board that is a bit larger than my usual size range and I was concerned the board cost might be getting up there so I checked with these guys. Not bad at all. But while exploring options I noticed some oddities.

When ENIG is selected, not only do they charge $16.00 for 'surface finish', but they add another $9 to the 'board' cost. So it's really $25 extra. Same with 2 oz copper weight. Odd.

Then when I checked the 1 layer price vs. the two layer price they lower the 'board' price just a bit less than $4, then add $4 to the 'engineering fee' so the total is a actually a little bit more!

The price goes up a lot with layers. They don't add much to the "board" fee, but the "engineering fee" goes up a LOT. I guess that's the labor and the board fee is the material.

But the prices are amazing! I had no idea.

Their web page is very clear and seems to cover almost every option. I don't see a way to specify the stackup as in layer thickness. They talk about fiducials being made as holes in the PWB either half way through or all the way through. Wouldn't this just be another drill hole? Or is this made with a higher accuracy?

Checking another board that is pretty small but 6 layer starts to get pretty expensive by the time I add in the smaller hole size and ENIG it's pushing toward $200. Still not bad really.

--

Rick C.

--- Get 1,000 miles of free Supercharging
--- Tesla referral code - https://ts.la/richard11209

 

[Lasse Langwadt Christense]

lørdag den 11. april 2020 kl. 18.22.46 UTC+2 skrev Ricky C:

On Saturday, April 11, 2020 at 11:19:17 AM UTC-4, Lasse Langwadt Christensen wrote:
lørdag den 11. april 2020 kl. 08.32.10 UTC+2 skrev Ricky C:

Aren't there pretty inexpensive ways of getting prototype quantities of PWBs made? I recall several outfits that panelize multiple designs into one batch so you can order a small number of boards at not a high price.


places like jlcpcb etc. will do 5x 100*100mm dual sided, with silk,mask etc.
so vastly better than anything you can make at home, for $5 the shipping might
be $10 these days

https://pcbshopper.com/

Someone is asking me to layout a board that is a bit larger than my usual size range and I was concerned the board cost might be getting up there so I checked with these guys. Not bad at all. But while exploring options I noticed some oddities.

When ENIG is selected, not only do they charge $16.00 for 'surface finish', but they add another $9 to the 'board' cost. So it's really $25 extra. Same with 2 oz copper weight. Odd.

Then when I checked the 1 layer price vs. the two layer price they lower the 'board' price just a bit less than $4, then add $4 to the 'engineering fee' so the total is a actually a little bit more!

The price goes up a lot with layers. They don't add much to the "board" fee, but the "engineering fee" goes up a LOT. I guess that's the labor and the board fee is the material.

But the prices are amazing! I had no idea.

Their web page is very clear and seems to cover almost every option. I don't see a way to specify the stackup as in layer thickness. They talk about fiducials being made as holes in the PWB either half way through or all the way through. Wouldn't this just be another drill hole? Or is this made with a higher accuracy?

at that price you don't get a choice of stackup, every one gets the same
because they nest everyones design on large panels, that's also why anything "special" like enig and 2oz copper adds up

I think fiducials half or all way through is for when you order a
solder paste stencil

Checking another board that is pretty small but 6 layer starts to get pretty expensive by the time I add in the smaller hole size and ENIG it's pushing toward $200. Still not bad really.

with jlcpcb you can also get vscored panel for not much more so you can
fit many smaller boards on the 100x100mm

 

[Lasse Langwadt Christense]

lørdag den 11. april 2020 kl. 18.53.38 UTC+2 skrev Ricky C:

On Saturday, April 11, 2020 at 12:44:46 PM UTC-4, mpm wrote:
On Saturday, April 11, 2020 at 12:22:46 PM UTC-4, Ricky C wrote:
On Saturday, April 11, 2020 at 11:19:17 AM UTC-4, Lasse Langwadt Christensen wrote:
lørdag den 11. april 2020 kl. 08.32.10 UTC+2 skrev Ricky C:

Aren't there pretty inexpensive ways of getting prototype quantities of PWBs made? I recall several outfits that panelize multiple designs into one batch so you can order a small number of boards at not a high price.


places like jlcpcb etc. will do 5x 100*100mm dual sided, with silk,mask etc.
so vastly better than anything you can make at home, for $5 the shipping might
be $10 these days

https://pcbshopper.com/

Someone is asking me to layout a board that is a bit larger than my usual size range and I was concerned the board cost might be getting up there so I checked with these guys. Not bad at all. But while exploring options I noticed some oddities.

When ENIG is selected, not only do they charge $16.00 for 'surface finish', but they add another $9 to the 'board' cost. So it's really $25 extra. Same with 2 oz copper weight. Odd.

Then when I checked the 1 layer price vs. the two layer price they lower the 'board' price just a bit less than $4, then add $4 to the 'engineering fee' so the total is a actually a little bit more!

The price goes up a lot with layers. They don't add much to the "board" fee, but the "engineering fee" goes up a LOT. I guess that's the labor and the board fee is the material.

But the prices are amazing! I had no idea.

Their web page is very clear and seems to cover almost every option. I don't see a way to specify the stackup as in layer thickness. They talk about fiducials being made as holes in the PWB either half way through or all the way through. Wouldn't this just be another drill hole? Or is this made with a higher accuracy?

Checking another board that is pretty small but 6 layer starts to get pretty expensive by the time I add in the smaller hole size and ENIG it's pushing toward $200. Still not bad really.

--

Rick C.

--- Get 1,000 miles of free Supercharging
--- Tesla referral code - https://ts.la/richard11209

Did you look at JLCPCB.com ?

We use them all the time (from China).
Great prices.

We have been buying mostly 2-layer, but we do buy some 4-layer occasionally (but not RF, which we prefer to order domestically)

Typical delivery is 5-8 days, but running more like 8-10 right now days with this Coronavirus thing.

Yes, jlcpb.com was the site. I bookmarked it.

https://lcsc.com/ is the same company so you can combine PCB and component order

They also have cheap assembly, though it is limited to a selection of LCSC
parts and only smd

 

On Saturday, April 11, 2020 at 3:21:43 AM UTC-4, Clifford Heath wrote:

On 11/4/20 4:10 pm, dagmargoodboat@yahoo.com wrote:
On Saturday, April 11, 2020 at 1:00:09 AM UTC-4, dagmarg...@yahoo.com wrote:
On Friday, April 10, 2020 at 8:19:13 PM UTC-4, Clifford Heath wrote:
On 9/4/20 9:39 am, dagmargoodboat@yahoo.com wrote:
I haven't been able to get toner transfer working since I changed
toner carts. But I've dabbled with two promising variations.

1) If you lacquer-coat the PCB *then* transfer the toner, the
lacquer fills in the toner's pores.
James,

What exactly is the stuff that Americans call lacquer? That's not a term
that has a single meaning here.

I'm not completely sure. I believe it might be shellac, made from
actual shells of actual lac bugs? The important property here is
that it's alcohol-soluble. You might be able to vet local products
by checking the clean-up solvents spec'd.

This is what I tried & recommend --
https://www.homedepot.com/p/Rust-Oleum-Specialty-11-oz-Gloss-White-Lacquer-Spray-Paint-1904830/100163605

I recommend white, as contrasting brilliantly with the black toner.

It's easily removed with 90% isopropyl alcohol, leaving printer toner
completely unaffected. Toner scoffs at IPA.


Acrylic, thanks. I had no idea that IPA would attack it but not lift
toner. That's good to know, I'll give it a try. I'll have to find
suitable local products, as the ones you mention don't ship to Australia.

I think you missed a note in there -- they've changed the product.
The original (that worked) was shellac(?), but it has been changed
to a new formula that allegedly does not work as well.

Ammonia strips acrylic floor wax, so maybe it would strip an acrylic
undercoating, if acrylic is all we can get.

I don't know which Rust-Oleum formula I got. Whatever's in my
two-or-three-years-old can of the stuff works wonderfully. It wipes
off the open surfaces readily with 90% alcohol(*), but stays in the
toner's pores, sealing them.

(*) 50% IPA was unsatisfactory.

I could try shellac, I have some flakes here. It's intrinsically more
variable though, being a natural product; the water content and age are
factors in using it for French polish. And it doesn't come in a
convenient spray can.

The toner image is printed on the paper you peel off the back of
adhesive labels, or in my case, backing peeled off adhesive
shelf-liner paper from the one-dollar store.

I tried things like that but they were too slippery - bits of toner just
fell off before getting transferred. I'm using toner transfer paper
bought for the purpose.

I had that experience early on, years ago. But the current label-
backing type paper didn't have that problem. It's truly a godsend;
no more rubbing, soaking and peeling, hoping the toner sticks.

The paper I use has a significant starch (I think) content, and wets
really quickly - like 10 seconds before it comes off cleanly. The only
other relevant factor is how much it slows down the heat transfer in the
laminator; even after proper pre-warming I pass it through 2 or 3 times.

Transferring to a lacquer-coated board might be easier. The lacquer
is infinitely grabbier than even roughed up copper.

Good to know.


BACKING PAPER
This gent illustrates my joyous experience printing onto backing paper,
and the ease of transfer--the toner just comes right off, eliminating
a wet step. He uses the paper he peels off Arlon vinyl.
https://www.youtube.com/watch?v=haqP8xhsYas

These are the labels whose baking paper is salvaged and used by
the gentleman in the first of those YouTubes above--

Best PRINT half-sheet labels (1/2 page-sized labels)
https://www.amazon.com/Best-Print-200-Half-Sheet/dp/B0069RY9BY

The big lure of backing paper to me was eliminating the process
variations of thermally transferring toner to copper from paper.
The backing paper goes a long way toward accomplishing that. I
still have trouble getting adhesion to bare copper.


Overall, this ability to use two different resist coatings (toner and
lacquer) that dissolve in different solvents feels full of
promise & possibilities...

Well, after all, this is only for Saturday afternoon prototyping at RF.

Yes, exactly. I've designed a class-C HF QRP PA using CMOS gates
that might see first power this way. LTSpice says it's about 80%
efficient, ideally, from d.c. input power to 50-ohm load.

Other prototypes can be on assorted grid boards that don't provide a
ground plane, and anything more permanent can go through one of the many
quick-turn board houses.

A friend uses a modified record-player to spin photoresist onto boards,
and makes double-sided boards quite frequently, but the process took a
lot of work to perfect.

Yep. The dry photo resist film method looks easier than spincoating,
but I'd really like to _reduce_ the wet steps and process variables.
https://www.instructables.com/id/Photoresist-Dry-Film-a-New-Method-of-Applying-It-t/

There's something about knocking up a schematic in Kicad, laying it on a
board, and soldering it up an hour after you started. Sometimes the
magic flows and you want to try something *now*.

There sure is something magical about it and that's exactly why I do
it. It's liberating.

Frankly I'd still be using unmodified toner transfer if it hadn't
quit on me. Previously I was making 0.012" traces and spaces with ease.
I had no problems at all, worked every time without trouble, and I
wondered what everyone else was griping about. But once it quit,
no amount of fiddling has gotten it working again. Same papers,
same printer, same clothes iron, etc., but no joy.

My problem is getting the toner to adhere to the copper. I had the same
difficulty transferring thermally, and with the cold transfer (solvent)
methods that rely on making the toner sticky--the darn toner just won't
stick to the board. But I must say toner sticks very aggressively to a
sticky lacquered board, whether I use heat or solvents to soften the
lacquer and make it sticky.

Cheers,
James Arthur

 

[Lasse Langwadt Christense]

lørdag den 11. april 2020 kl. 19.16.51 UTC+2 skrev dagmarg...@yahoo.com:

On Saturday, April 11, 2020 at 3:21:43 AM UTC-4, Clifford Heath wrote:
On 11/4/20 4:10 pm, dagmargoodboat@yahoo.com wrote:
On Saturday, April 11, 2020 at 1:00:09 AM UTC-4, dagmarg...@yahoo.com wrote:
On Friday, April 10, 2020 at 8:19:13 PM UTC-4, Clifford Heath wrote:
On 9/4/20 9:39 am, dagmargoodboat@yahoo.com wrote:
I haven't been able to get toner transfer working since I changed
toner carts. But I've dabbled with two promising variations.

1) If you lacquer-coat the PCB *then* transfer the toner, the
lacquer fills in the toner's pores.
James,

What exactly is the stuff that Americans call lacquer? That's not a term
that has a single meaning here.

I'm not completely sure. I believe it might be shellac, made from
actual shells of actual lac bugs? The important property here is
that it's alcohol-soluble. You might be able to vet local products
by checking the clean-up solvents spec'd.

This is what I tried & recommend --
https://www.homedepot.com/p/Rust-Oleum-Specialty-11-oz-Gloss-White-Lacquer-Spray-Paint-1904830/100163605

I recommend white, as contrasting brilliantly with the black toner.

It's easily removed with 90% isopropyl alcohol, leaving printer toner
completely unaffected. Toner scoffs at IPA.


Acrylic, thanks. I had no idea that IPA would attack it but not lift
toner. That's good to know, I'll give it a try. I'll have to find
suitable local products, as the ones you mention don't ship to Australia.

I think you missed a note in there -- they've changed the product.
The original (that worked) was shellac(?), but it has been changed
to a new formula that allegedly does not work as well.

Ammonia strips acrylic floor wax, so maybe it would strip an acrylic
undercoating, if acrylic is all we can get.

I don't know which Rust-Oleum formula I got. Whatever's in my
two-or-three-years-old can of the stuff works wonderfully. It wipes
off the open surfaces readily with 90% alcohol(*), but stays in the
toner's pores, sealing them.

(*) 50% IPA was unsatisfactory.

I could try shellac, I have some flakes here. It's intrinsically more
variable though, being a natural product; the water content and age are
factors in using it for French polish. And it doesn't come in a
convenient spray can.

The toner image is printed on the paper you peel off the back of
adhesive labels, or in my case, backing peeled off adhesive
shelf-liner paper from the one-dollar store.

I tried things like that but they were too slippery - bits of toner just
fell off before getting transferred. I'm using toner transfer paper
bought for the purpose.

I had that experience early on, years ago. But the current label-
backing type paper didn't have that problem. It's truly a godsend;
no more rubbing, soaking and peeling, hoping the toner sticks.

The paper I use has a significant starch (I think) content, and wets
really quickly - like 10 seconds before it comes off cleanly. The only
other relevant factor is how much it slows down the heat transfer in the
laminator; even after proper pre-warming I pass it through 2 or 3 times..

Transferring to a lacquer-coated board might be easier. The lacquer
is infinitely grabbier than even roughed up copper.

Good to know.


BACKING PAPER
This gent illustrates my joyous experience printing onto backing paper,
and the ease of transfer--the toner just comes right off, eliminating
a wet step. He uses the paper he peels off Arlon vinyl.
https://www.youtube.com/watch?v=haqP8xhsYas

These are the labels whose baking paper is salvaged and used by
the gentleman in the first of those YouTubes above--

Best PRINT half-sheet labels (1/2 page-sized labels)
https://www.amazon.com/Best-Print-200-Half-Sheet/dp/B0069RY9BY

The big lure of backing paper to me was eliminating the process
variations of thermally transferring toner to copper from paper.
The backing paper goes a long way toward accomplishing that. I
still have trouble getting adhesion to bare copper.


Overall, this ability to use two different resist coatings (toner and
lacquer) that dissolve in different solvents feels full of
promise & possibilities...

Well, after all, this is only for Saturday afternoon prototyping at RF.

Yes, exactly. I've designed a class-C HF QRP PA using CMOS gates
that might see first power this way. LTSpice says it's about 80%
efficient, ideally, from d.c. input power to 50-ohm load.

Other prototypes can be on assorted grid boards that don't provide a
ground plane, and anything more permanent can go through one of the many
quick-turn board houses.

A friend uses a modified record-player to spin photoresist onto boards,
and makes double-sided boards quite frequently, but the process took a
lot of work to perfect.

Yep. The dry photo resist film method looks easier than spincoating,
but I'd really like to _reduce_ the wet steps and process variables.
https://www.instructables.com/id/Photoresist-Dry-Film-a-New-Method-of-Applying-It-t/

There's something about knocking up a schematic in Kicad, laying it on a
board, and soldering it up an hour after you started. Sometimes the
magic flows and you want to try something *now*.

There sure is something magical about it and that's exactly why I do
it. It's liberating.

Frankly I'd still be using unmodified toner transfer if it hadn't
quit on me. Previously I was making 0.012" traces and spaces with ease.
I had no problems at all, worked every time without trouble, and I
wondered what everyone else was griping about. But once it quit,
no amount of fiddling has gotten it working again. Same papers,
same printer, same clothes iron, etc., but no joy.

My problem is getting the toner to adhere to the copper. I had the same
difficulty transferring thermally, and with the cold transfer (solvent)
methods that rely on making the toner sticky--the darn toner just won't
stick to the board. But I must say toner sticks very aggressively to a
sticky lacquered board, whether I use heat or solvents to soften the
lacquer and make it sticky.

tried a quick dip in etchant before the transfer?

a few seconds and the copper turns a more dull rough surface

 

[Ricky C]

On Saturday, April 11, 2020 at 12:44:46 PM UTC-4, mpm wrote:

On Saturday, April 11, 2020 at 12:22:46 PM UTC-4, Ricky C wrote:
On Saturday, April 11, 2020 at 11:19:17 AM UTC-4, Lasse Langwadt Christensen wrote:
lørdag den 11. april 2020 kl. 08.32.10 UTC+2 skrev Ricky C:

Aren't there pretty inexpensive ways of getting prototype quantities of PWBs made? I recall several outfits that panelize multiple designs into one batch so you can order a small number of boards at not a high price.


places like jlcpcb etc. will do 5x 100*100mm dual sided, with silk,mask etc.
so vastly better than anything you can make at home, for $5 the shipping might
be $10 these days

https://pcbshopper.com/

Someone is asking me to layout a board that is a bit larger than my usual size range and I was concerned the board cost might be getting up there so I checked with these guys. Not bad at all. But while exploring options I noticed some oddities.

When ENIG is selected, not only do they charge $16.00 for 'surface finish', but they add another $9 to the 'board' cost. So it's really $25 extra.. Same with 2 oz copper weight. Odd.

Then when I checked the 1 layer price vs. the two layer price they lower the 'board' price just a bit less than $4, then add $4 to the 'engineering fee' so the total is a actually a little bit more!

The price goes up a lot with layers. They don't add much to the "board" fee, but the "engineering fee" goes up a LOT. I guess that's the labor and the board fee is the material.

But the prices are amazing! I had no idea.

Their web page is very clear and seems to cover almost every option. I don't see a way to specify the stackup as in layer thickness. They talk about fiducials being made as holes in the PWB either half way through or all the way through. Wouldn't this just be another drill hole? Or is this made with a higher accuracy?

Checking another board that is pretty small but 6 layer starts to get pretty expensive by the time I add in the smaller hole size and ENIG it's pushing toward $200. Still not bad really.

--

Rick C.

--- Get 1,000 miles of free Supercharging
--- Tesla referral code - https://ts.la/richard11209

Did you look at JLCPCB.com ?

We use them all the time (from China).
Great prices.

We have been buying mostly 2-layer, but we do buy some 4-layer occasionally (but not RF, which we prefer to order domestically)

Typical delivery is 5-8 days, but running more like 8-10 right now days with this Coronavirus thing.

Yes, jlcpb.com was the site. I bookmarked it.

--

Rick C.

--+ Get 1,000 miles of free Supercharging
--+ Tesla referral code - https://ts.la/richard11209

 

[Jeff Liebermann]

On Sat, 11 Apr 2020 17:23:43 +1000, Clifford Heath
<no.spam@please.net> wrote:

Oh wow, quite a lot of things could be the problem. I better have a good
look inside before trying another cartridge.

Yep. Have an air compressor with a "blow gun" hose handy to clean out
the accumulated toner. About 30 psi seems right. The toner gets into
everything. I don't recall if the P2055 uses a laser or an LED array.
Probably laser. Sometimes, toner dust gets into the laser and
rotating mirror resulting in odd looking shadows. If you bypass the
shutter that protects the laser assembly, and blow some low pressure
<10 psi air inside, you might be able to clean it out without further
disassembly. Don't use the compressed air from a pressurized can.
Also, make sure the air dryer is working on your compressor so that it
doesn't spray rusty water or oil.

Once inside, inspect the rollers carefully, especially the fuser. If
you find melted blobs of toner all over the rollers, it's usually
easier to replace the rollers than to clean them. I made some tools
for scraping the melted toner off of the various fuser parts, but I
prefer to replace the entire fuser assembly rather than clean or
rebuild it. About $60 as I vaguely recall.

Putting a new toner cartridge into a dirty printer is a bad idea. It's
possible for the melted toner on the rollers to scratch the imaging
drum inside the toner cartridge. Clean first, then install a new
cartridge.

Old toner tends to clump together in lumps. This appears as uneven
coverage in large dark areas. It also damages the rubber wiper
blades. When the blades get old, the rubber hardens enlarging the gap
between the imaging drum and the blades. This dumps increasing
amounts of toner in the cartridge waste bin. When full, the waste
toner rubs against the imaging drum and scratches it. It is also
possible for toner "flakes" to stick to the imaging drum. That might
be another possible cause for the white spots.

05A carts are cheap on eBay:
<https://www.ebay.com/sch/i.html?_nkw=hp+05A+toner+cartridge>

Yet another resource:
<https://www.fixyourownprinter.com>

Thanks,
CH

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[mpm]

On Saturday, April 11, 2020 at 12:53:38 PM UTC-4, Ricky C wrote:

Yes, jlcpb.com was the site. I bookmarked it.

I didn't mention it earlier but we buy our RF boards (which are nearly always 6-layers and up) domestically because we often find that we need special stack-ups for impedance control (which JLC doesn't offer), AND lately on a few boards, the parts are getting so small and so densely packed that we need to tightly couple our Gerbers (silkscreen, pad tweaks, etc..) to the vendor's SMT line so the parts solder correctly during reflow, etc... We've had a couple cases lately where some tiny SAW filters (etc..) float off the (otherwise adequate) paste-up inspection. Resulting in re-work on our end to get the boards to meet spec.

All of which is a big wasted breath on the bean-counters in our company who simply can't understand why you can't shop the design around to just anybody who claims to stuff PCB's. They think they can just send the Gerbers and get perfect first article boards, no matter how complex (or high-frequency) the design is.

 

[Jasen Betts]

On 2020-04-11, dagmargoodboat@yahoo.com <dagmargoodboat@yahoo.com> wrote:

LACQUER
I found the technical data sheet, available on this page:
https://www.rustoleum.com/product-catalog/consumer-brands/specialty/lacquer-spray/

The operative info from that .PDF:
Resin Type: Acrylic Lacquer
Pigment Type: Proprietary
Solvents: Acetone, Methyl Ethyl Ketone, Toluene

It doesn't list alcohol as a solvent, but 90% IPA works like a beast.

alcohol is soluble in acrylic, and tends to soften acrylic paints.

UPDATE: apparently Rustoleum has changed the formulation of their
lacquer and it no longer dissolves in alcohol!!! The last 2 cans i
bought have been extremely difficult to work with, which might
explain some of the difficulty some of you were having duplicating
my results. I will update further with a substitute transfer base
soon. Thanks!
/quote

"Zinsser B-I-N" is shellac based and has an alcohol solvent, I have
purchased it in spray cans in the past. it contains a white pigment.

--
Jasen.

 

[Joe Gwinn]

On Tue, 7 Apr 2020 07:55:27 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

On 2020-04-06 21:35, jlarkin@highlandsniptechnology.com wrote:
On Mon, 6 Apr 2020 17:43:46 -0400, Phil Hobbs
[snip]

I guess the seive would soak up humidity and reduce the pressure
inside, so a tiny flow through the seals would introduce a little more
humidity. Wouldn't that eventually get to zero humidity and zero
pressure differential?

I guess atmospheric changes would still pump the system slightly.

Simon's the seal expert at this point. A sufficiently stiff box, with
enough screws holding the lid on, and hermetic connectors, ought to be
able to stay sealed pretty well. The pressure changes are nontrivial
though--our box is about 3 x 5 inches, so a 7% pressure change amounts
to about 15 pounds over the surface of the lid. They're also fairly
slow, so it doesn't take much of a leak rate to equalize the pressure.

We're using a cable gland rather than a hermetic connector, primarily
for cost reasons. I suspect that enough air will flow inside the cable
to manage the vent job, but we'll probably have to measure that to find out.

A traditional alternative is a long thin tube whose volume is
sufficient to ensure that no inside air gets out or outside air gets
in, despite the +/- 7% variation in ambient air pressure. Invented by
Louis Pasteur in 1859.

..<https://en.wikipedia.org/wiki/Swan_neck_flask>


Joe Gwinn

 

[Jasen Betts]

On 2020-04-11, Clifford Heath <no.spam@please.net> wrote:

Acrylic, thanks. I had no idea that IPA would attack it but not lift
toner. That's good to know, I'll give it a try. I'll have to find
suitable local products, as the ones you mention don't ship to Australia.

The "Export" brand paints available in Supercheap Auto are really
crappy, they claim to be enamel paint but seem to never actually cure
and always dissolves in mineral turpentine.

Supercheap also have 3M, Septone, and "SCA" brand "Acrylic" lacquer sprays.
As these paints are intended for automotive use they are probably alcohol
resistant due to the introduction of alcohol in fuel.

They also have SCA brand "multi purpose" Acrylic paint, this stuff will
possibly be softened by alcohol. Plasti-Kote and Dulux brand acrylic
spray paints from a hardware shop are probably worth a try too.

I could try shellac, I have some flakes here. It's intrinsically more
variable though, being a natural product; the water content and age are
factors in using it for French polish. And it doesn't come in a
convenient spray can.

"Zinsser B-I-N" is a commercialy produced shellac based primer. Spray cans
are available (but can be hard to find). Alternatively some places can put
paint into a spray can for you.

--
Jasen.

 

[Jasen Betts]

On 2020-04-11, Ricky C <gnuarm.deletethisbit@gmail.com> wrote:

On Saturday, April 11, 2020 at 1:00:09 AM UTC-4, dagmarg...@yahoo.com wrote:
On Friday, April 10, 2020 at 8:19:13 PM UTC-4, Clifford Heath wrote:
On 9/4/20 9:39 am, dagmargoodboat@yahoo.com wrote:
I haven't been able to get toner transfer working since I changed
toner carts. But I've dabbled with two promising variations.

1) If you lacquer-coat the PCB *then* transfer the toner, the
lacquer fills in the toner's pores.

James,

What exactly is the stuff that Americans call lacquer? That's not a term
that has a single meaning here.

I'm not completely sure. I believe it might be shellac, made from
actual shells of actual lac bugs?

I thought it was the shells too, but it's a secretion they leave on tree branches. So I guess they don't kill the bugs but harvest their secretions more like silkworms.

Actually, the silkworms don't survive harvest. Anyway it's a secretion from a
parasite, so its probably mostly resin... you could try pre-coating with liquid
flux or OSP.

--
Jasen.

 

[Lasse Langwadt Christense]

søndag den 12. april 2020 kl. 02.47.02 UTC+2 skrev pcdh...@gmail.com:

A traditional alternative is a long thin tube whose volume is
sufficient to ensure that no inside air gets out or outside air gets
in, despite the +/- 7% variation in ambient air pressure.  Invented by
Louis Pasteur in 1859.

Thanks, that's an interesting idea.(*) Riding on a big-ass harvester in E Texas ot Australia or some such place, it's fairly hard to ensure that rain or wash water never gets in, though.

Cheers

Phil Hobbs

(*) I have a lot of respect for the giants of the past--long ago, after I'd submitted my first and (so far) only theoretical paper, I withdrew it after finding that its main result had been published by Lord Rayleigh. One of my less merciful friends asked me, "How does it feel to be on the forefront of 19th century science?"

I seem to reading that probably around the same time period, there was a saying:
"If you think you have invented some new you need to learn to read German"

 

A traditional alternative is a long thin tube whose volume is
sufficient to ensure that no inside air gets out or outside air gets
in, despite the +/- 7% variation in ambient air pressure.  Invented by
Louis Pasteur in 1859.

Thanks, that's an interesting idea.(*) Riding on a big-ass harvester in E Texas ot Australia or some such place, it's fairly hard to ensure that rain or wash water never gets in, though.

Cheers

Phil Hobbs

(*) I have a lot of respect for the giants of the past--long ago, after I'd submitted my first and (so far) only theoretical paper, I withdrew it after finding that its main result had been published by Lord Rayleigh. One of my less merciful friends asked me, "How does it feel to be on the forefront of 19th century science?"

 

[boB]

On Sat, 4 Apr 2020 17:49:18 -0700 (PDT), pcdhobbs@gmail.com wrote:

Following up on blocher's sterling work,(*)

Many of us use parts off-label, often very successfully. A few examples:

SAV-551+ pHEMTs make very good wideband bootstraps. Their f_max is around 12 GHz, but they're amazingly stable.

74HC4352s make good flying-capacitor diff amp front ends.

TMUX1511s make very nice analogue lock-ins--their Coff*Ron FOM is almost in a class with relays, but 1E5 times faster.

Zero-ohm jumpers have about the right resistance to stabilize LDO regulators with ceramic output caps. (It's good to be able to disconnect the supplies during bring-up, and putting the jumper between the reg and the output cap has this additional benefit.

Your faves?

Cheers

Phil Hobbs
(*) who may be bulegoge's good twin, given the similarity of their emails

What a great subject !!

Kind of like off use of drugs for treating other diseases !

boB

 

On Saturday, April 11, 2020 at 1:29:23 PM UTC-4, Lasse Langwadt Christensen wrote:

lørdag den 11. april 2020 kl. 19.16.51 UTC+2 skrev dagmarg...@yahoo.com:
On Saturday, April 11, 2020 at 3:21:43 AM UTC-4, Clifford Heath wrote:
On 11/4/20 4:10 pm, dagmargoodboat@yahoo.com wrote:
On Saturday, April 11, 2020 at 1:00:09 AM UTC-4, dagmarg...@yahoo.com wrote:
On Friday, April 10, 2020 at 8:19:13 PM UTC-4, Clifford Heath wrote:
On 9/4/20 9:39 am, dagmargoodboat@yahoo.com wrote:
I haven't been able to get toner transfer working since I changed
toner carts. But I've dabbled with two promising variations.

1) If you lacquer-coat the PCB *then* transfer the toner, the
lacquer fills in the toner's pores.
James,

What exactly is the stuff that Americans call lacquer? That's not a term
that has a single meaning here.

I'm not completely sure. I believe it might be shellac, made from
actual shells of actual lac bugs? The important property here is
that it's alcohol-soluble. You might be able to vet local products
by checking the clean-up solvents spec'd.

This is what I tried & recommend --
https://www.homedepot.com/p/Rust-Oleum-Specialty-11-oz-Gloss-White-Lacquer-Spray-Paint-1904830/100163605

I recommend white, as contrasting brilliantly with the black toner..

It's easily removed with 90% isopropyl alcohol, leaving printer toner
completely unaffected. Toner scoffs at IPA.


Acrylic, thanks. I had no idea that IPA would attack it but not lift
toner. That's good to know, I'll give it a try. I'll have to find
suitable local products, as the ones you mention don't ship to Australia.

I think you missed a note in there -- they've changed the product.
The original (that worked) was shellac(?), but it has been changed
to a new formula that allegedly does not work as well.

Ammonia strips acrylic floor wax, so maybe it would strip an acrylic
undercoating, if acrylic is all we can get.

I don't know which Rust-Oleum formula I got. Whatever's in my
two-or-three-years-old can of the stuff works wonderfully. It wipes
off the open surfaces readily with 90% alcohol(*), but stays in the
toner's pores, sealing them.

(*) 50% IPA was unsatisfactory.

I could try shellac, I have some flakes here. It's intrinsically more
variable though, being a natural product; the water content and age are
factors in using it for French polish. And it doesn't come in a
convenient spray can.

The toner image is printed on the paper you peel off the back of
adhesive labels, or in my case, backing peeled off adhesive
shelf-liner paper from the one-dollar store.

I tried things like that but they were too slippery - bits of toner just
fell off before getting transferred. I'm using toner transfer paper
bought for the purpose.

I had that experience early on, years ago. But the current label-
backing type paper didn't have that problem. It's truly a godsend;
no more rubbing, soaking and peeling, hoping the toner sticks.

The paper I use has a significant starch (I think) content, and wets
really quickly - like 10 seconds before it comes off cleanly. The only
other relevant factor is how much it slows down the heat transfer in the
laminator; even after proper pre-warming I pass it through 2 or 3 times.

Transferring to a lacquer-coated board might be easier. The lacquer
is infinitely grabbier than even roughed up copper.

Good to know.


BACKING PAPER
This gent illustrates my joyous experience printing onto backing paper,
and the ease of transfer--the toner just comes right off, eliminating
a wet step. He uses the paper he peels off Arlon vinyl.
https://www.youtube.com/watch?v=haqP8xhsYas

These are the labels whose baking paper is salvaged and used by
the gentleman in the first of those YouTubes above--

Best PRINT half-sheet labels (1/2 page-sized labels)
https://www.amazon.com/Best-Print-200-Half-Sheet/dp/B0069RY9BY

The big lure of backing paper to me was eliminating the process
variations of thermally transferring toner to copper from paper.
The backing paper goes a long way toward accomplishing that. I
still have trouble getting adhesion to bare copper.


Overall, this ability to use two different resist coatings (toner and
lacquer) that dissolve in different solvents feels full of
promise & possibilities...

Well, after all, this is only for Saturday afternoon prototyping at RF.

Yes, exactly. I've designed a class-C HF QRP PA using CMOS gates
that might see first power this way. LTSpice says it's about 80%
efficient, ideally, from d.c. input power to 50-ohm load.

Other prototypes can be on assorted grid boards that don't provide a
ground plane, and anything more permanent can go through one of the many
quick-turn board houses.

A friend uses a modified record-player to spin photoresist onto boards,
and makes double-sided boards quite frequently, but the process took a
lot of work to perfect.

Yep. The dry photo resist film method looks easier than spincoating,
but I'd really like to _reduce_ the wet steps and process variables.
https://www.instructables.com/id/Photoresist-Dry-Film-a-New-Method-of-Applying-It-t/

There's something about knocking up a schematic in Kicad, laying it on a
board, and soldering it up an hour after you started. Sometimes the
magic flows and you want to try something *now*.

There sure is something magical about it and that's exactly why I do
it. It's liberating.

Frankly I'd still be using unmodified toner transfer if it hadn't
quit on me. Previously I was making 0.012" traces and spaces with ease..
I had no problems at all, worked every time without trouble, and I
wondered what everyone else was griping about. But once it quit,
no amount of fiddling has gotten it working again. Same papers,
same printer, same clothes iron, etc., but no joy.

My problem is getting the toner to adhere to the copper. I had the same
difficulty transferring thermally, and with the cold transfer (solvent)
methods that rely on making the toner sticky--the darn toner just won't
stick to the board. But I must say toner sticks very aggressively to a
sticky lacquered board, whether I use heat or solvents to soften the
lacquer and make it sticky.


tried a quick dip in etchant before the transfer?

a few seconds and the copper turns a more dull rough surface

Yes, thanks, I've tried a quick dip to roughen the copper surface,
but it didn't make any difference. It's mysterious.

When TT was working, I prepared the copper with fine steel wool,
then sometimes Comet (an abrasive kitchen cleaner with soap and bleach)
for de-greasing. I've the original method, rough sandpaper, fine
sandpaper, cross-hatching, and etching.

Cheers,
James Arthur

 

On Saturday, April 11, 2020 at 5:32:53 PM UTC-4, Jasen Betts wrote:

On 2020-04-11, dagmargoodboat@yahoo.com <dagmargoodboat@yahoo.com> wrote:

LACQUER
I found the technical data sheet, available on this page:
https://www.rustoleum.com/product-catalog/consumer-brands/specialty/lacquer-spray/

The operative info from that .PDF:
Resin Type: Acrylic Lacquer
Pigment Type: Proprietary
Solvents: Acetone, Methyl Ethyl Ketone, Toluene

It doesn't list alcohol as a solvent, but 90% IPA works like a beast.

alcohol is soluble in acrylic, and tends to soften acrylic paints.

UPDATE: apparently Rustoleum has changed the formulation of their
lacquer and it no longer dissolves in alcohol!!! The last 2 cans i
bought have been extremely difficult to work with, which might
explain some of the difficulty some of you were having duplicating
my results. I will update further with a substitute transfer base
soon. Thanks!
/quote

"Zinsser B-I-N" is shellac based and has an alcohol solvent, I have
purchased it in spray cans in the past. it contains a white pigment.

--
Jasen.

Thanks Jasen, very helpful. That should set Clifford on the right track.

Cheers,
James Arthur

 

[Gerhard Hoffmann]

Am 12.04.20 um 07:05 schrieb dagmargoodboat@yahoo.com:

On Saturday, April 11, 2020 at 1:29:23 PM UTC-4, Lasse Langwadt Christensen wrote:
(and others)

so much about problems and possible solutions that my mailer
would reject the post for excessive quoting..



The more I read here, the more I'm convinced to stay with the
photochemical priciple. The boards are cleaner than one could
ever hope to clean them oneself, surfaces lapped and covered
with photo resist and then protection foil.

All one needs is a UV source (face tanner in my case), NaOH
as developper and stripper, the etchant, Na- or Ammonium-
persulfate and two sheets of glass to enforce direct contact
between the film and the board during exposure.

I can easily produce anything my laser printer can print, easily
down to 4 mil / 4 mil feature size with offset film.
The process takes less than an hour from the time I've printed
the film and is absolutely reproducible.

I use their board material:
<
https://www.bungard.de/shop/index.php/de/fotobeschichtetes-basismaterial
>
but not their machines. There must be similar offers on other
continents.

And yes, I also use PCBway and others when there is time and
willingness to deal with DHL and our customs office
who question the price of the boards on a regular base.
That cheap, absolutely impossible! Fraud!

cheers, Gerhard