You are here: HomeThe SciEngDIY PCB Etch TankDIY PCB Etch Tank -- FR4 Single Sided PCB Test

DIY PCB Etch Tank -- FR4 Single Sided PCB Test

The last part of the PCB Etch Tank Series documented the technique I used to construct the tank. Well, I have finally been able to "test drive" the etch tank by transferring a test pattern onto a 203mm by 101mm cooper clad PCB using the press-n-peel blue toner transfer system. 


Figure 1: Single-Sided Copper Clad PCB etching in home made PCB Etch Tank. The etchant is sodium persulphate. The results have been incredibly encouraging as you can read below:


The aim of this exercise has been to test the quality of etching a PCB in a home made etch tank using sodium persulphate. A test pattern, Figure 2, has been created in a PCB layout package, with no real scientific significance, in an attempt to understand the limits to which the PCB etch tank can be used. Being able to successfully etch PCB traces that are 0.2mm in width could be important as it would mean that 0.5mm pitch quad flat pack devices could be used on home made PCBs. 


Figure 2: The PCB test pattern created in Protel DXP. The signal traces increase in thickness from 0.2mm to 1.0mm in 0.1mm increments. The gap between signals traces is 6.5mm. 


A description of the method used can be viewed in the Figures below.


Figure 3: The test pattern has been transferred onto the copper clad PCB. The test pattern has been transferred to the PCB using the blue press 'n' peel transfer system  and a dry iron. Holes have been drilled at opposite ends. Knitting wool (it was all that was available at short notice!)  has been passed through the holes to suspend the PCB in the etchant. The PCB approximately measures 203mm by 102mm and is 1.6mm thick.  


Figure 4 : Preparing the Sodium Persulphate. (a) An empty container has been weighed at 36.4g and since 100g of sodium persulphate should be dissolved for every 500ml  of etching solution the intention was to weigh 400g of etchant to make 2l of etchant solution. To my surprise the 500g (Net weight) bottle contained approximately 401.8g so I decided to use the lot. Hmmm. 


Figure 5: Preparing the Etchant Solution. The 400gm of etchant has been added to 2l litres of water at a temperature of 12.6o Celsius (a). Quite worryingly the sodium persulphate immediately congregated at the bottom of the etch tank (c). The 50W heater has been immersed such that it is below the "waterline" as the temperature sensor of the heater, used to regulate it, appears to be at the top of the heater (b). The external temperature monitor has been placed such that it is roughly halfway between the bottom of the etch tank and top of the "waterline" of the etchant solution. (Some websites suggest adding the sodium persulphate to warm water however I am not sure whether this method is more appropriate than adding the sodium persulphate to cold water as according to the label on the sodium persulphate container the solution rapidly degrades above 50o Celsius - B.P). 



The temperature vs time measurements can be viewed in the graph below. The temperature measurements have been taken by observing the reading of the digital thermometer on the side of the etch tank approximately every 5 minutes. It appears the digital thermometer's  probe has not been affected by the etchant. 


Figure 6: Etching Temperature vs Etching Time Measurements. This graph shows the results of measuring the temperature at approximately five minute intervals during the etching process. The PCB has been inserted into the etchant solution at approximately 37.5 oCelsius after the etchant solution had been warmed for approximately 1 hour and 50 minutes. It took approximately 30 minutes to completely etch the board at a temperature of about 40 oCelsius which is less than the temperature at which the best results could be achieved of 45 oCelsius. The troughs or drop in temperature that appear in the graph occur when the top of the etch tank has been removed to mix the solution. It appears that the chemical reaction between the sodium persulphate etching solution and the cooper on the PCB releases energy in the form of heat which probably accounts for the increase in temperature after the PCB is inserted into the etchant solution. 




Figure 7: The Copper Clad PCB is inserted into the etchant solution at about approximately 37.4 oCelsius. 



Figure 8: Copper Clad PCB after 12 minutes (39.4 oCelsius). All seems to be going well although it is immediately noticeable that copper has etched quite rapidly close to the heater and close to where a lot of air bubbles are escaping from the airline. It seems that the oxygen contained within the air bubbles could be aiding the etching process too.  


Figure 9: Copper Clad PCB after 17 minutes (39.2 oCelsius). Work in progress.


Figure 10: Copper Clad PCB after 22 minutes (39.2-39.4 oCelsius). Almost all of the copper has been etched apart from where the PCB is close to objects like the airline. This would suggest that the airline should be eventually tethered to the bottom of the etch tank. Note that the etchant solution is becoming progressively more blue as the chemical reaction that removes the copper from the PCB takes place. 


Figure 11: Copper Clad PCB after 27 minutes (40.2-40.4 oCelsius). Well, the final result looks spectacular although after I had removed the PCB I noticed a few specks of copper that could potentially cause short circuits in a real circuit. Perhaps I should have left the PCB in the solution for about another 5 minutes.


Figure 12: The Empty PCB Etch Tank After Etching . I suppose that the etchant solution should become more and more blue as more copper clad boards are etched. 



This  has been a great test run that fills one with nothing but confidence. It might be worth buying another heater to place in the opposite side of the etch tank especially if double sided PCB's are to be etched. The next thing to do is layout a TPS61165 test circuit, that could be double-sided, as another test for the home-made PCB etch tank. Well done me! 


Figure 13: The Final Result. Although the etch resist has not yet been removed and there are some specks of copper that have not been etched the result is marvellous. The potassium sulphate etchant solution should last for approximately 6 - 8 weeks before it needs changing so expect quite a few experimental PCBs between now and then. 

Go to comments start