Hi. This video comes about because of a post that uh Simon one of the evev blog Forum moderators made on the Forum asking a question of what is the typical Uh pla in thickness of a PCB VR and that led on to you how much current you can put through and stuff like that. And it's an interesting question. And as it so happens, my little micro ruler actually has a handy little table on it a very small one for uh, the amps for a typical 1 mil plated uh VR for various temperature rises versus uh, the Via diameter and well, that's the answer.

Rough rulle of thumb is that a typical PCB via is going to be about 1 mil thickness. Um, that's 1 mil as in 1 th. it's not 1 millimeter, 1,000 of an inch or around about 025 mm thickness. And but it's going to vary a lot depending on the manufacturing.

uh, the manufacturer, the manufacturing process, and the manufacturing tolerances. But you know that's a rough rule of thumb and I've always taken as a PCB designer. A nice safe value for how much uh, an individual VR can carry one single. VR is rough rule of thumb about 0.5 amps.

so if you want uh, a trace to go from one side of the board to the other side of the board, it's got to carry two amps. Uh, you typically put at least four vs in there to be on the safe side, but it's a very interesting question and it also brings up uh, the question of uh, can you just uh, plug or fill the VR with uh solder to increase the current carrying capacity and well, this comes back to the videos which I'll linking down below that both Mike from Mike's electric stuff and I have done on uh proving that uh sold decoding PCB traces does actually make it a difference. Not a huge difference, but it does make you know a significant difference to the Uh in decrease in the resistance of a PCB trace and I was wondering is that the same for a VR So I thought I'd actually try it and also uh, get out the Uh Via calculator and plot some graphs and do that and we don't know what we're going to get. Could be interesting so let's have a quick look at it.

So the first thing I'm going to do is use one of the these Uh calculators. This one is, uh, an awesome one. If you haven't got it, you definitely should get it. It's the Satin PCB uh toolkit calculator and uh, it has all sorts of you know stuff.

I'm sure I've shown this on the Uh blog before, but if we go into Uh Via properties here, then we can actually set uh, multi-layer boards micro vas like this. but uh, what we want is the two layer board standard uh 1, Point uh 6 mm thickness. It gives you the 1.57 5 and then we can set our whole diameter like this and then our pla in thickness. We've got set to a, you know, sort of a rough standard uh 1 mil.

That's the default thing which is 1 mil of course, um 1 th which is 0254 MM plating thickness and then down here say for a 1 mm diameter hole it we can solve that and then it gives us the Via current down here 1.12 amps and the Uh via resistance as well are 0.33 milliom. Now, we don't care about the inductance and stuff like that cuz we're only dealing with Uh DC here. and then of course, um, you can calculate from the resistance and the current. You can calculate the power of course and it gives you the cross-sectional area and it gives you the thermal resistance down here and the voltage drop.

and you know, uh, it's really quite nice, but that is for a 1 C temperature rise and that's the thing. Uh, when you're designing a board like this, you have to uh, determine what is your acceptable temperature rise in that copper PCB Trace usually or in this case, the Uh VR which is basically exactly the same, except it has a less controlled manufacturing process than a uh, a regular trace on a PCB and uh, you know a default value might be a temperature rise of Uh 10 C But you know quite frankly, when I'm designing boards, you know, belt and braces type stuff I don't want to piss away 10 See all that power in my uh traces? Unless I'm doing real high power stuff right on the edge and you know I'm the physical limits of how big I can make traces and stuff like that? then I'm going to set that temperature rise for 1 C So basically you know, very conservative I don't want you know almost any power at all dissipated in the VR like for example, here we go where we've got 0.4 for 1 mm diameter hole 1 C temperature rise we've got4 M there dissipated in our Via. Fine if we set that order of magnitude bigger to 10 C temperature rise. look at that 4 M God you can fly halfway to the moon on 4 Ms You got to be kidding me.

So um, what I'm going to do is uh well, what I have done is uh done. uh all these values uh, whole diameter St starting at 0.1 mm all the way up to 2 mm for two different temperature rises 1 C and 5 C like that and I've gotten the values out of this and t-a I've plotted them in a spreadsheet here and this is the graph we get. It's very interesting. Check it out.

it's uh for as I said for the 1 mil uh plating the one th pla in of your via which is going to be fairly typical but it's going to vary a lot depending on the manufacturing process and that's the thing with these Vas. This is just going to be rough rule of thumb stuff and does our uh, rough rule of thumb that I've known of. you know, roughly 0.5 amps um per V to be on the safe side and well, let's have a look here. on the vertical axes, on the leth hand vertical axes I've got the Uh the calculated current capacity which the toolkit uh told me here and that's from 0 to 4 amps up here and on the right hand vertical axis we've got the VR DC resistance in milliohms.

So these two traces here correspond to the Uh VR DC resistance and these two traces this. this series here and this one here are for as you can see: the red is for the 1 C temperature rise there and the green is for the 5 C temperature rise. and uh, it's a very interesting result and check it out. It's basically linear um for the current capacity versus whole diameter there, but they are a couple of little kinks.

You can see a A like, you know little kinks in there. It's not. It's not absolutely completely linear. so I don't know where the satin PCB calculator is getting its um, you know how it's actually doing its calculations.

It's obviously getting it from the IPC 21 uh standard somewhere and taking into account thermal uh, you know properties of copper and all sorts of stuff like that. So I'm not exactly sure why we're getting that Kink there at 1.6 um, 1.5 1.6 amps it drops back down and then continues up and I've got that. The same is for the 1 C temperature rise and the 5 C temperature rise. Now you'll notice that the slope of the 1 C uh temperature rise one here is of course lower than the 5 degre C temperature rise.

And if you found and if you plotted 10 C you'd probably get a steeper uh ramp there yet again. So I haven't plotted that one in I Don't know where it would, uh actually end, but there you go. you can see it's effectively pretty linear with or it appears to be Um according to the calculations that the Satin PB calculator is using, fairly linear with respect to whole diameter. So you double your whole diameter, roughly double your current cap capacity here.

Now do we get our rough 0.5 amp? Uh. Rule of thumb: Well, let's take the 1 C temperature rise. which is you know if you're going to design your board, why have it for 5 or 10 see temperature rising your copper? That's just stupid. So one is a good figure which I always uh, tend to use unless um, otherwise and look here it is I'm in the smallest uh diameter you're going to be using on a standard drilled diameter on a standard board is about 0.3 mm.

That's sort of one of my standard VR small VR uh whole sizes and you know it. There it is. You know we're roughly getting you know, 6.7 maybe? uh, current capacity for one of those Vas. So that's that.

Rule of thumb pretty much holds true that rough to be on the conservative side. No, anything under this curve 0.5 is under that curve for all diameters. So really yeah, that's confirmed. And if you have a look at the Uh VR DC resistance here, you can see that it's not a linear Uh curve as you'd expect.

that is curving up right, tails up right drastically. right at the bottom end. Sort of. You know, under that 0.5 mm, it really starts to tail up like that fairly drastically.

So there's not a direct Uh linear correlation there between the Uh current capacity and the DC resistance of the VR. So that is certainly a very interesting graph and it would be fascinating to uh, get some graphs of some other calculators as well. I Mean this is just the Satin PCB calculator. Uh, specifically.

and I'm sure it's going to vary depending on the type of calculator and what calculation it's using to estimate that current capacity. but I Think this one's probably not a bad rule of thumb at all. I Rather like it, all right. Well, let's do a practical test and see if we can actually measure a difference on a typical Uh VR or hole on a PC pad, hole on a PCB.

There is no difference practical difference between a VR and a pad. They're just used for different Uh things, really, but they're essentially they're constructed exactly the same way on a blank PCB so we don't necessarily have to use an actual VR. We can use a component pad which is exactly what we're going to do here and we see if we can measure a difference in the resistance of this VR after we actually played it through. now.

I'm not going to, um, pick one at random here because we actually need a special case and I found a board. We need a four uh terminal measurement because I'm going to be passing current through the VR and I don't want it to go anywhere else I want all the current to flow througha and then we want to be able to tap off voltages. So I've rummaged through some old Uh boards and it was actually quite difficult to find a board with exactly the configuration of tracks which I needed. Which is this, um, this pad here is the one that we're going to actually Target and I'm going to feed the current in here and it goes down there and it can go nowhere else.

Because this Trace is open down here, it can go nowhere else but through that via or pad there and then out this side. So that's it. And the good thing is it's got two tap points. They all sense points directly on there.

so I can attach a wire to here and it will sense the voltage on that surface surface of the pad there. and I can also attach a wire to here and that will sense the voltage directly at that pad there. So that should be pretty good. So we're going to use a four terminal uh resistance measurement, but we're not going to use of course the four terminal measurement on our bench multimeter cuz it only measures at one current.

we want to measure a much higher currents. so I'll show you the test set up in a minute. but this one should work quite well. And this is yes, this is a solder coded board so those Vs are actually ated through and solder coated so they're not bare copper and they're not gold.

So if you wanted to do this methodically, you would try uh this with not only different size uh vs and holes but for each type as well. for Bare copper plus uh, solder uh, solder coded ones or you know, solder coded ones like this and also uh, gold. uh, flash plated ones as well. But we're just going to start out with this solder one to see if we can measure the difference.

so we'll measure the resistance of that pad before and after we fill it with solder, see if we can measure the difference. I'm pretty confident we will be able to. and in case you're wondering, this is an old board. This is a Um Hydrophone calibration test box.

Jeez, when's Thompson Marone Sona Jeez, those were the days. Don't even don't know if I got a date code on there, but hey, there's Dlj with a smiley face up there. This was a dual channel uh charge amplifier so you can tell Op Amp Arrangement here with the Dual uh dual capacitors in there and this was in a charge two channel charge amplifier configuration for the DT, the device under test, and the reference, and then a power little power amplifier on here to drive a speaker. Sort of a self-contained Uh calibration test box.

Anyway, let's get to it. and I don't have the original files to check the exact whole size here, but uh, roughly. based on my little hole gauge here, it seems to be about 1.2 mm. Not that the actual Uh Di diameter is going to matter, we just want to be able to see if we can see a difference.

That's all we're going for here. The absolute value doesn't matter, just want to see if it actually changes. So my test set up here is my Ryol uh power supply set to uh constant current mode I can just program in the current you can see I've probably got. yeah, you can likely see that.

1 amp, uh, constant current programmed in there at the moment and then I'm reading uh, the voltage across directly across the VR on my Uh 6 and2 digit Adant bench meter and there you go. you can see it's 655 MTS all thereabouts and I'll show you the four terminal measurement down here. So as I said before, current flowing into this pad down here and you might be able to see that on the bottom of the board anyway. it flows around there on the bottom, up through that via there, and to the other side of the power supply.

So there's our current Loop there. so it's a direct short on the output of the power supply. of course, the power supply uh, puts a constant current through that and then I'm tapping off a sense line there. That's the negative sense line and this positive uh, sense line goes on the bottom side as we saw before to the other side of the pad there.

So we sensing directly on that pad or via there. and the reason I don't actually attach the wire directly to there is because there's no drop across here at all. It doesn't matter, Um, effectively in infinite input, impedance of the multimeter makes no difference whatsoever. and I don't want to disturb that pad either.

I Want Um, I Don't want to like Reflow the solder joint before and after. So now we've got a control condition where we're measuring the exact voltage across, uh, that pad, the resistance of that pad, and we know the current flowing through it. So Ohms law. We can calculate the resistance of that pad.

Now all we need to do is uh I'll do this for different currents 0.1 amps all the way up to 2 amps so we can actually get a graph of this thing and uh, then I'll do it before I'll do all that data before and after I fill that with solder and then with wire. So as we saw before, 651 MTS I'm just going to ignore that last digit. You know it's that's really bit meaningless due to noise, but this will be more than good enough to uh, uh, check I think and um, with uh, precisely almost precisely one amp flowing through that um OHS law tells us we're looking at 65 MMS in that uh particular pad there. But as I said, absolute value doesn't matter.

We want to see if that changes and of course I can go in here and I can just go. uh, not point. Oh yeah, select uh, not 1 amp and we should see that drop by an order of magnitude to 065. There we go.

It certainly does. It's linear and you can see that with no current flowing through there, we're getting a bit of an offset. I've actually got. You know it's physically disconnected down there, so what we're going to do is just null out that value.

Oh, there we go. We' nuled that out and we can now take our readings and I can see a slight creeping up. I'm now at Uh 2 Amp So this is the highest I'm going to go. 1310 MTS is slightly going up.

that's due to the Uh heating up of the actual VR down in there. And let's see if we can just do a crude temperature measurement on this thing with my thermocouple ambient here in the lab. 26.4 de, uh, 26.5 de C Let's call it that. it's going up a bit.

maybe cuz I've got my hand a little bit close to there. Anyway, let's do a Delta on that and see if we can get a physical measurement. I Got to press down it. Really, You know it's not the most accurate.

If you had a nice thermal camera of a really good close-up macro lens, you might be able to get it, but we're getting about a two I don't know I could leave it there for a bit Jiggle It Around try and get the maximum, you know 2.2 de C Let's call it that and I Just went to my Satin Calculator program just to verify what you would get for a 1.2 mm uh pad. typical 1 mil plated I Don't know if this board is mil PL but it's got some extra solder coat on there whatever. Near enough for a 2.2 de C uh, temperature rise Delta on that pad and what maximum current. It worked out to about 1.8 which is pretty close to the 2 amps that we putting through.

so we're certainly in the ballpark. It's all working out. So I've got all my data from 0.1 amps up to 2 amps in .1 amp. uh, increments, 20 data points.

more than enough and they're the voltages I get. we can plot that, that's for the Be pad. So I'll do this again for the solder coded plot. But first of all, let's do a live test, see if it changes.

Now for the big test. we're currently 1312 MTS and what I'm going to do is just fill this pad with solder. I'm going to do it. Live here and we should see if the theory is right.

Should see this value drop. That's if the solder codeing through that vaa makes a difference. This will confirm or deny the myth. The myth is that solder makes no difference whatsoever To to that you have to put a wire through.

Well, we'll test the wire thing in a minute, but let's see if it makes a difference I think it will. It won't be large, but we should see this value here drop. now. Um, we expect it to initially go up because I'm going to heat up the pad.

but uh, once it cool down, it could take like a minute to really cool down. But let's try it. Here we go. Okay, I'm on the pad, it's going up.

Yep, going to apply some solder. So what was it? 1.31 Something like that? There we go. I think it's flowing all the way through. We'll have to check.

Aha, Look, it does make a difference. There we go, it hasn't. It's still cooling down. There you go.

That is quite a significant difference. Myth: Busted that solder doesn't make a difference. It does fairly significant actually. And the solder has kind of come all the way through.

But just to be sure, I'll just coat the top side of that pad as well. So it's got a nice, beautiful, even uh, code of solder flowed all the way through standard 6040 solder. and uh, we'll let that cool and take our measurements. And there is the data for the bear pad and the solder fill pad and look at that.

It dropped from 1309 to 779. almost half. Wow, Of course this. I Expect this to vary a lot, depending on the you know, the manufactur, the manufacturer of the board, the uh, differences in manufacturing, that process, tolerances, and all sorts of stuff for VAs and holes.

There, you know there is quite a lot of, uh, manufacturing tolerance in these things, but on this particular Solder Field board, certainly look at that. a dramatic difference. Now you could say that the other myth is that, well, uh, because the solder in the pad doesn't make a difference. You have to solder a wire through the pad if you want to, or the Via if you want to, increase the current capacity of that via.

Well, let's try it. I got a big ass resistor here. Um, that leads you know, like 1 mm diameter or something, right? Big chunky lead. So I'm going to solder that through.

Uh, live here. and by the way, I can solder live? It doesn't matter if this solder in Iron's Earth or not because everything's floating here. So I am able to solder on live equipment. Just be aware of that if you are doing uh tests like this just to make sure it's all floating and all your solder and iron isn't earthed.

Anyway, let's do it live. I'm at 2 amps. Uh, there we go. 0.777 Will it make as big a difference? I.E Will it half again? I I I don't know I don't know I've never actually tried it.

Oh well, you know I've done it. I've never actually done a controlled experiment on this. There we go. I've sold it through.

There we go should be solded nicely on both sides and uh, yep, yes it has made a difference. It has made a difference. and in fact, look at that. Yeah, it's made of substant I Didn't think it would make my gut feeling was it wouldn't make a huge amount of difference.

but it yeah it certainly has. but uh the you know. But the solder also made a big difference so that one is certainly confirmed. Uh, you do get uh, lower resistance by soldering a uh copper wire through the pad like that, but because of course, the copper is not directly touching the copper outer walls of the uh pad or ver itself, you know it's not going to make as huge a difference as it would if that hole or V was actually filled with solid copper to begin with.

And there's all our data. Tada to the spreadsheet and here we go: I've entered the Uh data in the tables here and but that's just the voltages versus the current. I've converted that into resistance down here. and then I've plotted all this data and taada.

Look at our graph. uh, linear as you'd expect a little bit of tailing up, a little bit of nonlinearity at the bottom end there. but uh, basically linear from 0.1 amps all the way up to 2 amps and look at that dramatic drop. It's like I don't know the exact figure, maybe at least 40% Um, there, you know.

It's like quite a significant drop just from. so this is the bare vea to the solder plugged VR So just filling it with solder, we definitely busted that myth. Um, it does quite significantly and drastically reduce the resistance of that via and hence increase its current uh handling capability for a given temperature rise and then, um, once again, it halves yet again for the wire. So if you go from the solder coat to the wire there, what is it? Maybe a third or something like that.

but yeah, there you go. Um, so the wire one is confirmed. Yes, that does make a difference and but the solder one does too. and uh, well.

I Haven't seen any Um data on this before I Don't think there's probably something out there, but um, yeah, this is quite interesting bit of uh research here. I'd love to actually do it further and do all the different types of Vs with the different whole diameters and get a whole parametric curves and something like that. That would be fantastic. That would be a nice little Um experiment to do.

like a bare. just a bare via with just uh, you know, with no gold coding, no solder coding. so just the bare copper VR and then the Uh solder cated VR as we did here and then also a goldplated one for example. or maybe a silverplated Uh one for example and that would be most interesting.

But as I said, think the data is going to vary a lot depending on the actual Uh PCB and the manufacturing process used because there's a lot of Tolerance very large tolerances. It's not like 5 10% you know we could be looking at, you know, 100% Um, you know something like that difference in, you know, like doubling of values, stuff like that for uh, various Uh manufacturers in their manufacturing process, technology and what they guarantee in terms of their Uh plating, minimum plating thickness and stuff like that I have no idea what that board was I used the plating thickness used in there, you know but it did work out to basically very similar as I said to that uh value which I've got here if I have a look at it there it is um it worked out to that 1.8 amps down here that for that 2.2 de temperature rise which we actually uh measured and uh with hindsight it's not really surprising that the value actually dropped filling the hole with solder because as I've linked in uh, previous video showing that it does make a difference on traces on a PCB if you coat them in sold it not nearly as much as uh it does seem to here at least in this particular um example because the top of a PCB is a fairly uniform thickness whereas I've just picked this uh page off random off the internet PCB 007.com thank you very much. Um, the this shows a typical double-sided plate through hole and you know these Corners here. um where the copper attaches, you know, very thin and this is much thinner of course than the consist and it's less consistent than the thickness of the copper on the top.

so if it makes a difference on the top, imagine what the variability in the corners in the connection in the corners down there are like that and when you then have a nice big solder fillet on top and then going all the way through, you'd expect it uh, to make a larger difference than it does if you just sold a coat a very consistent amount of copper on the top of the board here and this one actually shows a copper uh, wire going through as we experimented with there and of course you notice that it's not touching either side whereas if it was, if it was like J like jammed in there, then you Pro probably would lower it a lot more because you'd have direct copper contact with sort of that side and upper surface and the corners there. So if you really jammed it in there, but putting the wire in there does Make a Difference by lowering the resistance, but still, there's no direct contact so you got to go through the actual solder to get from the copper on one side to Copper to the other to lower that resistance. but it still does make a significant difference so it all works out. I Found that really interesting I Love getting data like that and plotting it.

and uh, actually measuring so I Hope we busted a couple of myth there. Yes, you can just solder vas. Fantastic. Who knew? There you go? Well I hope that's cleared that up and you found that interesting.

If you do want to discuss it, jump on over to the Eev blog. Forum Catch you next time.