Yogi rides the new 6 wheeled robotic base platform. Designed in Solidworks and milled on my CNC machine.

Tuesday February 21 , 2017
Font Size

PCB Through-hole at Home


I figured out a way (after hearing some rumors it was possible) to through plate vias and thru holes using a liquid from a "car defroster repair kit" sold by permatex.com. I found it at Advanced Auto Parts for the cost of about 10$ US. You can also use "Silver Print" made by MG Chemicals (see update below).

My pcb had 10 mil traces and via holes using a #72 drill. Many of the vias worked. Normal through holes like 2.54mm headers also worked, even some large holes worked. Typically the through holes were the best as they were big enough to clear the liquid through without getting clogged. However, with the vacuum table, it should suck the #72 vias fine.

First Attempt - Promising!

For my first attempt, after etching both sides of the pcb, but before drilling, I placed a large sticky label over the pcb (mine were transparent like scotch tape). I then setup my cnc machine and drilled all the holes. Using the conductive liquid from the defroster repair kit and an old glossy business card I squeegeed the liquid across the PCB over all the drill holes. I then used a vacuum on the pcb bottom side to suck the liquid through the holes. Though overall the conductive liquid stays liquid for a long time, it dries quite fast once it is spread thin. It reminds me of non-newtonian liquids such as corn starch mixed with water. This first attempt worked for about 40% of my vias but typically the liquid dried before I could get the vacuum to suck it all the way through the hole to complete the connection. Clearly, this method requires a vacuum table to suck the liquid through while it is being squeegeed.

My pcb had 10 mil traces and via holes using a #72 drill. Many of the vias worked. Normal through holes like 2.54mm headers also worked, even some large holes worked. Typically the through holes were the best as they were big enough to clear the liquid through without getting clogged. However, with the vacuum table, it should suck the #72 vias fine.

The sticker you place on the top and bottom of the PCB is of course to protect the top and bottom of the PCB from getting any conductive liquid on it. How conductive is it? Very conductive! The paper I was using as a drop cloth got soaked with the liquid and at about 5 inches the resistance was 0.002 ohms. The vias turned out to be about .015ohms, perhaps the paper had more of the liquid? Tin coating the board afterwards will probably lower the through plating impedance as well. Using the clear Avery labels I used are probably not the best choice, they are plastic based and don't drill clean but microscopically tear. Paper labels or perhaps painter's tape will work better.

Attempt 2 - Vacuum Table  - 100% Success!

For my second attempt I built a vacuum table out of a 4$ cutting board. With my CNC machine I milled a graph pattern deep (75%) into the board. This allowed the PCB to sit on any part of the cut-out and still have lots of air circulate underneath. The graph cut-out was about 4x5 inches, I would tape over any part of the vacuum table not covered by the PCB to get a good seal so air would only be sucked through the holes in the PCB. I used plastic plumbing pipe for 4$, some sheet gasket, and sheet metal to secure an ingress vacuum port that fit my household vacuum cleaner. I tried using a compressor and two different air bed pumps too. A compressor supplies a lot of pressure but cannot provide the volume of air we need to create a good vacuum at the pcb holes. The air bed pumps were close, would probably be adequate, but the household vacuum is plenty of volume and pressure and so it the right tool for this job. The vacuum cleaner could hold the pcb to the vacuum table with quite a bit of force!

The Vacuum Table

The vacuum table with the vacuum attached and covered with painters tape to size the vacuum area to the PCB size. The basic vacuum table made using my CNC table

The vacuum table is made to suck the conductive liquid down through the PCB. (Click to zoom.)The table contains raised nibs so I can support small sized PCBs by sealing around the PCB with tape. Tape is used to seal areas around the PCB and the vacuum table to ensure vacuum air only through the PCB holes for maxumium amount of pressure through the PCB holes. Shown above I have blue painters tape to the size of the PCB (not shown). The small orange strips will raise the pcb slightly off the table so no holes are blocked by any part of the table.

Preparing the PCB

I used blue painter's tape to cover the pcb before drilling. The tape drilled very nicely. I am lucky to have a CNC machine that drills all the PCB holes. I only need to know the locations of two distant holes for machine alignment and therefor I don't require a full drilling template. If you don't have a CNC machine and are drilling by hand you should use paper Avery labels and print your drill or gerber artwork to use as a drill template. Since the PCB and paper is somewhat transparent held up to light, it is quite easy to align the template.

After drilling the holes I placed the PCB on the vacuum table and taped it down with more painter's tape creating a seal so the only air ingress (intake) is through the PCB holes. If you have large holes, like mounting holes, cover these with a small piece of tape while you squeegee the small holes, then remove the tape and finish the mounting holes seperately. I did about three squeegee passes for this attempt, but I probably only need 1, maybe 2. After finishing and removing the PCB from the vacuum table I placed the PCB into the toaster and baked. This will remove any remaining water from the conductive liquid and solidify it. (I should have removed the protective painter's tape first.)

There was much liquid that was sucked through and fell into the vacuum table. Clearly 3 squeegee passes was too much. I used acetate to clean the vacuum board and this worked to remove it all with very little scrubbing. It may also be possible to reconstitute this extra liquid, for a small amount of this liquid is expensive!

The results for this attempt were 100%! I only had 9 holes in this small pcb, but they were with a #78 drill (0.019in), so very small. Since I baked the board with the painters tape, I burned the tape and the adhesive became very sticky and difficult to remove from the PCB. Next time I will remove it before baking for sure! I scrubbed the sticker off with a straight edge of another pcb, a razor blade and an abrasive scotch-brite pad. The amount of abrasive scrubbing I did I am amazed the conductivity was not broken, or that water didn't reliquify the paste but after a long process of cleaning the board I still had 100% conduction! I just proved the results of this method are not flimsy. I could have used acetate to remove the adhesive but this is obviously death to the conductive liquid too. Luckily I discovered that before trying to clean the board!

Update: After visiting the permatex.com website and looking at the Material Safety Datasheet I discovered that in fact the secondary ingrediate after the dissolved silver compound is in fact acetate. So acetate would redisolve the conductuve liquid but you should also be able to reconstitute left over material by cleaning with acetate and then letting the acetate evaporate to thicken the liquid again.

Other sources of conductive liquids

Update: I am trying to track down other less expensive sources of conductive liquid. One source is from MG Chemicals called "Silver Print". The price of this stuff is around 40$ for a 1/2 Troy Ounce. This is considerably less expensive than buying in the small amount each defroster repair kit contains. This liquid works well but takes a while to dry and does not become conductive until it dries completely. Once you are done spreading this liquid in the holes, you should leave the board to sit and dry before pulling the protective tape off the board.

If you know of inexpensive sources or a way to make your own, please contact me.


* The liquid from the defogger repair kit drys in mere seconds, too fast, when it is spread thin...so by the time I suck it with the vacuum its usually already pretty solid. Some holes seem to suck it through via capillary action (?).
* Though you can do a few holes at a time, probably better to put a good amount on and squeegee the entire board at once, the liquid stays liquid when thick so it spreads well and you dont have to rush.
* During the first attempt, some holes were clogged with liquid, some were clear. More often the clear ones work, the clogged ones don't as the liquid does not reach the bottom copper.

Problems with Conductive Fluid (new)

I've had a problem with the Silver conductive fluid recently. I used a few vias to tie the ground of my regulator to the ground plane on the bottom of the PCB. In operation, this regulator's ground pin had a tendancy to float on occasion. Perhaps it had to do with current output, I am not sure. The resistance of these vias would intermittently change upwards to 50ohms! Perhaps this plating is sufficient for small signal traces, but I am hesitent to depend on them on vias for power distribution.

There are a few other things I've noticed about the Silver fluid that suggests to me that the copper fluid is better. The silver solution takes a long time to dry and until it does it is not conductive. The copper solution dries almost too fast, but it is also almost instantly conductive. I will try the copper solution on the next board for sure.

Making the PCB

I used the toner transfer method to create the PCB. I just bought the transfer paper, the TRF foil and the TIA laminator from the guy at www.pulsarprofx.com. This method is excellent! It is so much better transfer than regular iron and tonor...the green TRF foil applied over the tonor is very resistant to abrasion and the etching process. This method can easily etch 8mil, they claim it can do 6mil with practice. Also, buy the TIA laminator for 60$, it's so worth it. My regular iron couldn't supply enough heat or pressure. The TIA lays down perfect tracks with full toner transfer and no broken traces.

I've used the photo developing method in the past. It's not as easy as the toner transfer above but I could produce very good results with practice. At the time, pulsar's toner transfer method was not available and the toner transfer method's that existed were poor for surface mount parts. One thing I learned about the photo method was rather than using transparencies, I had a Litho company produce litho transparencies for me for about 10$ (circa 1996). Regular transparencies had a lot of UV leakage within the black areas and timing the exposure was difficult. These lithos are used in exposure equipment for developing metal plates for printing presses. The exposure bulbs in this equipment are *very very* bright and so the black portions of the litho's must be super black with no leakage. With these litho's I could expose the board with UV for an hour and still not expose any part under the black. 3 minutes was more than adequate, but the point is timing the exposure was not an issue any longer.

Beyond the exposure problem, alignment of the top and bottom layer was the only thing I could mess up. Alignment still remains a difficult part of DIY PCBs when you are dealing with small via sizes like 20mil vias with 12mil holes. I hope to come up with a perfect alignment method soon too.


Comments (17)
Rivets as an alternate solution
17 Monday, 26 December 2011 11:50
Hi C,
Just thought to let everyone here know of also another method that I found. Not sure if you are aware of it.
It is using rivets.

Here is the link:

Home made vias: The hobbyist's Pursuit for perfection ;-)
Just one more YouTube Link
16 Sunday, 25 December 2011 12:43
Just thought to share this link with everyone.


Take care.
An alternative product?
15 Sunday, 25 December 2011 12:25
Hi. Just thought to mention another product that might be usable, and also cheaper than the MG item. I found the MG Silver here in Oz from the importers to be sold at about $65, and the currency being at parity at the moment, that is way to much.
Here is what I found, it is called Silver Conductive Varnish.
Here is the link:
The price has remained unchanged for the last 4 years.

I have read somewhere that they are ok for signal but not for heat transfer, but that was for RadioShack conductive pens, and things and most of those posts that I found references to were very old.
Having searched more, it is a small vial of 3gr.

I am also looking in to the copper too :-)
Wish you all the best. I have bookmarked the page to be able to be checking bak on how things are progressing and if there have also been any other ideas.
14 Sunday, 25 December 2011 00:39
Hi Colin, Yes you are correct. I did not realise how minute of an amount it is.
Do you think we can come up with another idea, to be able to re-use that small amount again, without it being contaminated? Like a smaller power vacuum or pump, that can go under the board but have layer of some type of slightly porus paper, so it can be sucking air through it, but capturing the solution so it can be used again? How is its viscosity, is it very thick or slightly liquidy? Just curious as I have not used any of this yet.
I have sent a message with my email address.
13 Saturday, 24 December 2011 23:59
I saw a video in youtube in which an ink with particles of graphite is used as a conductive medium before electroplating. It surely would be less expensive than silver, and plated would be reliable too. Maybe even not plated, but I'm not sure, never tried XD
Using pump to draw conductive fluid up
12 Saturday, 24 December 2011 14:24
Aries: Ah, I see what you are saying now. However, I dont think you realize how expensive this conductive fluid is - like gold. It costs me $25 for an amount the size of an almond. It's better to squeegee a small amount around the board and suck it through with air. (btw: Dont forget to leave your email address in the form so I can email you back.)
11 Saturday, 24 December 2011 03:47
Hi C,

Here is a quick and simple drawing I did using Paint :-)


All that will need to be done is moving the hose above the board as to be sucking the fluid from underneath so it can coat the via / hole.

Having had a brief look at pumps, they are around $40 depending on the strength.

Again, this is just a theory that it should produce consistant and quick resaults.

p.s. I hope that the drawing is not to primitive ;-)
Reply to: Hi Aries
10 Saturday, 24 December 2011 03:04
Hi C,

Just a normal water pump or for fluid, search on ebay for an aquarium water pump or for ponds that circulate the water or from your harware store for a water pump, then place the plastic tube, as you do in your version for normal air, but this one, always wearing gloves ;-) just move it over the board to get the fluid going thru the vias and the return you put in the tub as to be recycling the fluid.

I hope I made it clearer for you.
Google the following:

water pump
aquarium water pump
no need for the filter, as you don't need to filter the liquid. All you are diong is just circulating the fluid with vacuum.
Hi Aries
9 Saturday, 24 December 2011 02:37
Thanks for the response. What would the "liquid pump" do? Is it conductive? I haven’t heard of this liquid.

The etching process I do is very good. It takes only a little bit of etchant and etches very fast - less than 8 mins for a double-sided 6in x 4in board. I place the etchant and board in a ziplock bag, then that bag into another ziplock. Then I place this into a larger bucket of very hot water. Hot enough I can barely keep my hand in. By the way, this hot water doesn’t mix with the etchant, it's just to warm it. I then use a rubber roller in the hot water to roll across the zip lock bag - squeegeeing the etchant back and forth across the board. Very easy, no mess, and easy cleanup. It helps to get as much air out of the sealed ziplock bag as possible.

The low etch times makes for nice clean trace lines. I can get 6mil detail easy if I want to.

I also use the pulsarprofx system to put the artwork on the board. I've used all the methods (including photo developing) and this is my favorite and easiest with best results.

All my boards are just test units, so it's populated tested...then usually thrown out when I get the boards professionally done.

Liquid Pump
8 Saturday, 24 December 2011 02:25
Just a thought mate. Have you ever considered on using just a small plastic container or pyrex. Fill it just above the pcb size, so the pcb is totally immersed, use a "small" fluid aka liquid pump and have the pcb in the tupperware or pyrex and run the pump on top of the vias / holes with the out going back in the container. That way, it gets pulled thru the holes and is fully immersed.

Then do the silver / tin etc plating with Liquid Tin or Silver.

I have not tried it yet, due to funds and just starting to look at different ideas, and when I came accross your site, it just gave me this idea.

A pet shop should have a pump for aquirium use, and also a heater for the ferric chloride for etching, if that is how you etch. Just remember never to use the heater for an aquarium ;-)

I will have a look at the rest of your site now too. Very nice article.
Clogged holes often fail
7 Saturday, 04 December 2010 03:20
Hi Jason, initially I did this before making the small vaccum table. I found that holes that clogged often werent conductive where holes that were clear were basically always conductive. It may have to do with the surface tension of the liquid, or van der Waals forces, I'm not sure. But it's probable that in many cases the two sides of liquid dont "come together" but are both stuck as a coincident plane to the pcb surface. The vacuum ensures that the liquid comes all the way through and once a slight layer goes through then more easily follows.
6 Friday, 03 December 2010 23:50
Why not simply fill the VIA and let it clog then turn over and fill that side to make a solid VIA which both sides touch?
Plated Vias
5 Thursday, 12 August 2010 15:09
Here is my solution: http://DiyPcbVias.com
4 Sunday, 27 June 2010 12:37
My friend, that text is very ispiring. Knowing that this kind of operation can be done at garage keeps me trying to search and use techniques I leaned.
Think and Tinker
3 Sunday, 06 September 2009 19:04
I saw those pages too. I plan to try the electroplating soon, but I am in for the long haul. I think making an electroplating station (and sourcing the chemicals) is too much for many PCB hobbyists. Also, I want to try aluminum anodizing too so I can kill two birds with one stone.
Plated through-holes
2 Sunday, 06 September 2009 15:13
Michael B.
Why don't you try copper-plating the through holes after you've activated them with your silver ink? Doing this will result in much more robust and reliable vias, and it's not all that complicated to do. This site gives a good overview of the process:

Hole activation in pcb
1 Wednesday, 17 December 2008 19:37
Did you ever tried 'mechanical' activation? Is a method made with placing the pcb above a conductive solution and drill it on the cnc. The drill bit will go down the liquid and place conductive material in the hole when exit. It is mentioned on IPC manuals but just mentioned not any more information. I think it could be done with some sponge material full with a silver solution, but will be better to drill and clean the board first to avoid all the dust. But all this work should be followed by copper plating. A serious plating line the chemicals will go to 600USD the activation is based on palladium witch it self is very expensive. But there are about other ways to activate the holes. Good work!

Add your comment

Your name:
Your email:
Your website:
User Rating: / 170

RSS Feeds

Visitor Poll

What sort of peripherals do you desire in a robotics main board? (you may vote more than once.)

Who's Online

We have 64 guests online