Hi In a recent uh, tear down video power supply tear down I mentioned um that a cheap, simple, and very common way to um uh, decrease the resistance and increase the current handling capability of PCB traces is to leave the solder mask off and uh, let the wave solder apply extra solder to the trace like this and uh, there were quite a few uh people who said well, that's really not going to make a huge difference at all. um if anything, well and uh I I I think they're wrong. It's a common technique that's been used for a long time, but you know I don't know if anyone's actually done the numbers on it so I thought I'd do a video actually uh trying to measure the difference. unfortunately Mike at Mike's electric stuff has beat me to it.

he's already done a video uh checking just this. He uh did some measurements uh before and after he took off. um the solder actually sold Wicked it off and uh did some measurements I think he got a figure of like 40 to 50 uh% um uh, decrease in the resistance of the Uh Trace due to the Uh solder on there which is quite significant. um you know, so it it definitely does make a difference as you'd expect.

uh but I uh noticed an issue with uh Mike's videoing that obviously um he started with a board like this. already had the Uh solder on it and as you can see it's it's can be quite inconsistent. This process um you know some like in some areas it'll be big and lumpy, others will be very thinly coated like that so you know it's a very hit and miss uh thing. but it is a cheap and simple way to do it anyway.

Mike Uh started with a board like this with the solder already on it and then went to Wick it off and remove it all and um I think that's you know, a back to front way to do it and uh can incre and can probably um have significant Uh error involved in there because I don't think there's any way you can possibly solder Wick off all of the Uh solder on there and be left with your original copper. Because copper is only standard copper on a board like this is only uh 1 oz uh copper and uh, that is 35 microns thick. It's very very thin so I thought the correct way to do it is to uh or a more accurate way so we can get some more measurement data is to start off with a standard 1 oz copper clad board with a single trace of a known Uh thickness like this and uh, then take some measurements of that of course and then add the solder to it and see what the change is and that should give us um, more accurate uh readings than what Mike got. So let's give it a go.

I've got some uh farell uh verab board here and it's uh part number 4789 and I had to look up the uh MSDS actually material safety data sheet to get the thickness. but it is standard 1 o 35 Micron copper. So let's give it a go. And as I mentioned in uh, my power supply video and uh Mike mentioned as well, there are other ways uh to do it apart from add solder to the tracers, how to get increased current handling capability Um, in your tracers, that's to use thicker copper, even 2 oz or even 4 oz copper.

But the issue with that is a it's very expensive to do and uh, B uh, if you've got a mix of very high power power supply stuff on one part of your board and very, uh, dense stuff on another part of your board, say you've got a BGA or you know, very fine pitch quad flat pack or something like that. Um, then etching away 4 oz copper. Um, you know, to those sort of tolerances required for very fine traces can be very, uh, difficult, if not impossible. so you can't sort of mix and match high power stuff with high density stuff using that thick copper.

So this is a real cheap and uh, simple way to do it. And of course, that's the reason most manufacturers will do this. Uh, very common in power supplies is because it's cheap, simple, and you'll effectively get it for free. Now of course, to do these measurements properly, you got to use what's called four wire resistance measurement.

Um, and I've uh, done this in previous videos, you've seen it and uh, we'll do uh, two different Uh methods of four wire. We'll use my HP 3478a and then we'll also, um, do the manual method of Uh passing the constant current through and measuring the voltage drop with a multimeter just to get you know, two different measurements to make sure we're consistent. So a four wire measurement basically means that uh, you have a drive wire which is this one and then you have a sense wire as well. So that, um, when it's driving current through these leads, you're going to get a voltage drop and an error along these leads.

But if you have this sense wire which is tapped right at the same point like that, then you're measuring the voltage directly off there. And because there's no current, you've got a high impedance multimeter measuring this. There's no voltage drop along this wire. It measures the exact uh res the exact voltage drop across your Trace like that.

So it is the most accurate way to measure resistance like this. And of course we've got that on both ends of the board like that. So we got one at that end and one at this end. four wires.

We can drive it with a constant Uh current or with a a four- wire multimetal like a HP 34 78a. So we've got two Drive wires and two sense wires. Let's give it a go. And just for the record, the length here from Uh Cents point to cense point is ah 358 mm or thereabouts.

The thickness is approximately 4.2 mm. And for those curious to know the difference between four wire and two wire mode, if we switch to two wire, we get much higher 220 uh, 212 milliohms because all of the uh, it's measuring all this wire in series with it. but you switch to four wire and it effectively cancels out any resistance in these wires and the contacts and things like that. Actually, that's come down a bit.

Maybe due to, uh, you know, it was still a bit warm from the soldering, perhaps. Anyway, 51 milliohms and what? I've got here: I'm now passing a constant current of 1 amp through this Uh trace and I'm measuring the voltage drop from the sense wise and as you can see, 51 M Volts for 1 amp use Ohms law, that's 51 milliohms. It's spot on to our HP 3478a meter. Uh, four wire terminal resistance measurement.

So we've got two different Uh measurements confirmed This Trace is definitely 51 milliohms. Now we can add some solder, see what the difference is, and the solder I'm using today will be standard multicore brand 6040. None of this lead free rubbish, just your traditional 6040. So let's give it a go.

Um, I'm not going to go all the way right to the end, right to the tip there because I don't want to, you know, upset my uh, measurement, upset my connection there. So I'll go most of the way and I'll put a very thin coat on to begin with. So let's give it a go. This will be a prob you know, a thin and then I'll spread it along and you'll see the resistance jump all over the place due to uh, once these uh, copper heats up, it changes its resistance of course and so does the solder.

So uh, really, we need to let it settle down. but we've got 51 momms there so let's let's give it a go. Well, that's rather surprising. It's not, uh, nearly as much as I uh, thought it would be.

It's a relatively thin, uh coating on here, but I get a uh, basically it's dropped to 43. 2, 43.3 milliohms and and uh, I calculate that as basically a 15% decrease in resistance and uh, uh, or a 177% increase if you want to. U do the way Mike did it and I checked Mike's numbers again and he got a 40% increase so mine's only 177% He got 40 so he must have had a lot more solder on there I can only presume or he got a 28% decrease. I'm getting a 17, uh, 15% decrease, but it's still.

it's um, relatively significant considering that you get it for free like you don't. Actually, you know you have to. All you have to do is leave your sold mask off and it's H. You certainly wouldn't rely on the fact uh, cuz it wouldn't be consistent.

But anyway, I think what I'm going to do is put another layer on, make it really thick and globby and uh, see what we get and it's uh, it's it's actually cooled down. by the way. it's as you can see, it's pretty stable. It took about, you know, a few minutes to cool down at least.

So let's apply some more solder and you can see my coating there. It is relatively, uh, thin. It pro maybe looks a bit thicker than what? I on camera perhaps. but yeah, I let's uh, let's add it.

let's go much clumpier like down here. So I've started to add some. let's do some more. Even clumpier than that I think All right now we're talking 26.2 milliohms so you know we're basically that's like a 94% uh, increase in the resistance if you're talking in the way Mike did it or a 48% uh, decrease in resistance practically? you know, double and half.

So and really I don't think my solder is as thick as I'm not sure how that looks on camera, but really, I'm not sure it's nearly as thick as what you know has accumulated on this board after the wave soldering process. And you know that's that's fairly typical of a board after. it's uh, you know been, uh, uh, wave solded like that in 10 So you know I think yeah, it's basically um, we talking. You know you have the resistance.

um, pretty much so. that's a better result than what Mike got much better. He only got a um, a 40% increase in resistance or a 28% decrease. I'm getting a 50% decrease and I know what you're thinking.

can we get it back to normal and do what Mik did and remove all the solder and get it back to 51 milliom. Well, let's give it a try and that's basically what's left after uh, trying to Wick it all off I could probably get a little bit better than that, but gee, you know you could do this thing until the cows come home, but there's yeah. Trust me, there's really not much uh solder left on that at all in terms of uh, Phys IAL height. Anyway, okay, I've done my dest to, uh, suck all that uh solder off the strip and we're back to 52.

MMS or thereabouts going to be a little bit of error in there, but and I may have even. um, you know, Phys CU I physically scraped along with the stuff, so maybe it actually takes off. You know, a little bit of uh, copper in there or something. you know.

Maybe it even leeches it. uh out. Maybe it even leeches a bit of copper out I don't know the chemical, uh process precisely. But anyway, we're basically back to where we were.

so um, that pretty much validates uh Mike's technique of starting out and removing the solder as well. But I think our one was a little bit more, uh, accurate because we did start off with the noan uh quantity with the 1 oz copper. So there you have it tinning a PCB Trace We've done some I think reasonably controlled measurements here, so you can be pretty confident of these results. We got anywhere from a 15% to 50% decrease, a half in of the resistance by just 10 in the trace with 60/40 lead solder it.

It'll differ with the tin stuff, but that's a nice ballpark. That 50% uh figure might got a figure of 28% uh, decrease in resistance. We were able to get, uh, better than that. So there you go.

It's going to be somewhere within that range, so it does actually make a difference. And considering that you can do this for free by just removing the solder mask and getting the uh solder you know, put on there by The Wave soldar in process, it's You know it's not a bad technique. it's um, and that's why people have used it for a long time. It's cheap.

Well, it's effectively free so and it does work. But you can't rely on it though because as you see, there is a quite a big spread in there. And uh, even if you leave a very thin coat of solder on there, it, it really doesn't uh, do much at all. You've got to have at least a, you know, a reasonable mount on there when you try and suck it all off.

It basically goes back to the same as the copper, but that's very interesting result. I Don't know if anyone's actually done that before, so it's it. Pretty much uh, confirms Mike's results as well. So there you have it, some conclusive results and a bit of an industry rule of thumb there.

I like it. If you want to discuss it, jump on over to the Evev blog forum and if you liked it, please give it a thumbs up. Catch you next time Sh!.