Chemical pump for the PCB etchant tank


I'm ramping up on building a new circuit board prototype, so I figured it was time to come up with a better pump for my etchant tank. I've used bubblers of various sorts, but the best results to date have been obtained by using a pump to flow the Ferric Chloride etchant across the surface of the copper clad.

I've tried various pumps and have had the best luck with the $5 surplus windshield washer pumps. The only drawback is that after 4 or 5 boards, the acid eats through the metal shaft on the pump and that is the end of it.

I've searched all over for a cheapo pump that was chemical resistant, but I have not been successful and finding such an item that was small and cost effective. I must admit that sometimes I forget that I have a CNC mill in my garage that is capable of milling complex parts. So, I set out to build my own acid-resistant chemical pump that would fit the bill of my PCB etchant tank. Here is what I came up with.

UPDATE: After using the chemical pump on the first circuit board, the results were great, but the pump is just too powerful for this application. I ended up getting Ferric Chloride all over the place because the pump was just making a swirling vortex of etchant and it splashed quite a bit. I was quite impressed with how even and consistent the etch came out though. So, this pump will probably find use in a different application that requires a higher GPM than the etchant tank requires. I'm now starting to think that a magnetic stirrer is the way to go because the only thing that touches the acid is the stirring stick. I'll have to hit Ebay and see if I can find a magnetic stirrer/hot plate combo for a reasonable price.

The chemical pump in action!


DXF cad drawing of the proposed soulution


After cutting the parts from black Delrin stock, I chucked up the pump housing and finish cut the pump cavity to size.


The impeller was almost perfect right off the CNC mill, so all I had to do was bore the hole to size for the plastic impeller shaft.


The test setup I used worked quite well. I still have to measure the head pressure and the GPH, but I'll do that once I get the motor mount finished.


The intake side of the pump.


Here you can see how the pump goes together. Everything fits quite tight and there is minimal play in the bore and cavity to impeller clearance.


With the impeller assembly removed, you can see how I turned a piece of the black Delrin to a shaft which mates nicely with the impeller. Everything is friction fit and should stay together until pressed out.


Impeller and shaft assembly.


I drilled the pump outlet at a tangent to the impeller cavity. I really wasn't sure if there was any science to this, so I just went for it and it turned out quite nice.


The entire assembly opened up. Once I got the impeller to the correct thickness, I ran the fly-cutter across the pump housing to get the clearances as close as possible so the pressure would remain adequate. You can see some of the flycutter marks on the housing.


Motor mounts are next! I don't know how fast this pump was spinning, but when I tested it out, it sucked itself onto the bottom of the tank. I'll probably have to add a filter and some feet to keep the inlet off the bottom.