LCR Cal Test PCB

[Martin72]

I always like to use these golden support points for my PCBs, they are suitable as test pins as well as wiring support points.
Narrow design but still very stable.
Edit: Pic added.

[KungFuJosh]

This is the new kelvin clip connection test point I made in Kicad. The opening is 1.8mm x 4mm with a 2mm gap between openings. I'll keep it this size if it fits, otherwise I'll drop it down a little.

Thanks,
Josh

[Martin72]

I wonder how long something like this can last.

[shabaz]

optionally use the board with VNAs too.

I'm curious, as I've never had a VNA. What about this board would be useful to VNA users?

It's almost the same thing as the LCR meter, but the implementation is different internally. Unlike an LCR meter, one downside is that the sweet-spot for accuracy is far narrower, but on the positive side, the measurement frequency range can be large (to GHz and beyond if desired, although the minimum frequency is often say 10 kHz or 50 kHz, it would be rare for it to function down to say 50/60 Hz, although some VNAs can be made to operate to such low frequencies).

Instead of Kelvin clips, there are just two-wire connections using coax usually. The way it works is that the VNA causes an AC voltage to be applied at a frequency, and some of the signal reflects back from the device-under-test if it doesn't look like exactly 50 ohm at the given frequency, and the VNA measures the reflection and compares it internally with what was sent (in terms of signal level and phase), to make its internal calculations. After a frequency sweep, the output can be plotted in various selectable forms.

The photo shows a DIY board that is similar to your test board, but only with RF connectors and no kelvin clip holes. My board has the open and short (as @mawyatt mentions), but for a VNA a 50-ohm termination is also needed for the calibration, whereas LCR meters often don't need that. If your board was used with a VNA (for say measurements from tens of kHz to tens of MHz), then a 50 ohm resistor location (could be 49.9 ohm, or two 100 ohm in parallel) would be good.

My board only has space for one component that I have to desolder if I want to measure a different component. Eventually the SMD pad can tear off due to the repeated soldering operations, so it isn't that good unfortunately.

I made this board by copying an existing TI board Gerber files (it was a reference design for PCB antennas, but I deleted the antennas from the Gerber files and copied the 'open' portion to make the DUT area (with the solder mask removed in the Gerber files from the end of course).

The chart shows example output for a particular inductor.

[KungFuJosh]

I wonder how long something like this can last.

If made with a good copper weight/ENIG, I'm sure it would last fine. The connection holes can be cleaned and a thin layer of solder added eventually if needed. If you drag the kelvin clips out tight and hard a few hundred times, that might be a problem. If you're gentle, I don't see why it would ever be an issue. I probably won't use it even 50 times. Not all of use have 30 LCRs. 😉

[Martin72]

I will use my test pins.

That leaves the question of the components.
Tight tolerance resistors are a dime a dozen, with capacitors it's a different story (2%), with coils the same (5%).

[KungFuJosh]

It's almost the same thing as the LCR meter, but the implementation is different internally.

Thank you for the explanation. Between your explanation, and your example, I don't think my board would be great for that. A modified version of it might though.

How many components/values do you want to test? I dunno if that crazy plated area on your board is needed or not, but if it is, a multi-component board could get quite large. If it's not needed, then I would need to know the values and component types that would be useful on a VNA variant of the test board.

[KungFuJosh]

I will use my test pins.

lol, fair enough. Perhaps I'll make a hybrid connector version so people can easily make that choice. What size mounting hole do you need for those cute little pins?

[KungFuJosh]

Here it is with 1mm holes, I'll adjust to what size you need of you like the idea.

[Martin72]

1mm could fit, but I'll measure it again tomorrow.

[mawyatt]

I always like to use these golden support points for my PCBs, they are suitable as test pins as well as wiring support points.
Narrow design but still very stable.
Edit: Pic added.

That image with the Blue components looks like a Twin T Notch Filter?

Best,

[Martin72]

Jepp,
I had designed it according to your idea, for the Bodeplot, if you still remember it.

[mawyatt]

Short term memory failing fast

However, saw the 4 Blue Caps, and 4 resistors, thinking the Blue caps look equal size/value (can't tell about Rs tho), so what passive network (didn't see any active stuff) could this be??

Should produce nice notch if the Caps are matched assuming Rs are typical 1%.

Best,

[Martin72]

I can show it again in the 3000X HD thread, but not here.

[shabaz]

Thank you for the explanation. Between your explanation, and your example, I don't think my board would be great for that. A modified version of it might though.

How many components/values do you want to test? I dunno if that crazy plated area on your board is needed or not, but if it is, a multi-component board could get quite large. If it's not needed, then I would need to know the values and component types that would be useful on a VNA variant of the test board.

That board is intended for up to a GHz, but realistically for measurement up to some tens of MHz,  if not a lot of accuracy is wanted, then even a solderless breadboard is fine. Often users don't mind if there's (say) 5-10% error, because inductors and capacitors have tolerances anyway, and some circuits can trim that out. The photo shows a board that comes with some NanoVNAs, used for through-hole components.
Personally I don't need a lot of component positions, just as long as 0603 up to 1210 or so components can fit. But no need to design specifically for these requirements, I was just thinking out loud in case it was useful for you or others too. Maybe it's a bit too niche to add to your existing layout, unless you're planning to explore with a VNA at some point.

[KungFuJosh]

Ha, I bought some of those boards, built them, and forgot I had them. I hate 0603, I try to avoid them. I'm too old and wear glasses. 🥸

[tautech]

I will use my test pins.

lol, fair enough. Perhaps I'll make a hybrid connector version so people can easily make that choice. What size mounting hole do you need for those cute little pins?

After spotting the reply from shabaz, another thought for pins came to mind.

TH turned IC pins we sometimes use instead of IC sockets < plentiful and cheap.

[KungFuJosh]

I'll keep it simple and make the hybrid option I shared above. I just need confirmation on the hole and pad sizes. That way anybody can do what they like.

[KungFuJosh]

I took a look at these pins from keyelco: https://www.keyelco.com/userAssets/file/M65p104.pdf

Seems like a 1.5mm hole is better, so this is what I'm going with unless somebody has strong feelings about it:

[mawyatt]

How are you going to arrange these thru component holes so you can use the Kelvin Clips on each end of DUT. If the DUT is a leaded component then you'll need two holes for it that are separated by the DUT spacing, then at each end/hole you'll need the Force and Sense holes for the Kelvin Clip.

Needs to look and work something like our sketch earlier.

Best,

[mendip_discovery]

That leaves the question of the components.
Tight tolerance resistors are a dime a dozen, with capacitors it's a different story (2%), with coils the same (5%).

I have often wondered this, as I don't think I want to go to the effort of the large standard resistors but at the same time I would like some with good properties that mean consistency and little drift over time.

[KungFuJosh]

How are you going to arrange these thru component holes so you can use the Kelvin Clips on each end of DUT. If the DUT is a leaded component then you'll need two holes for it that are separated by the DUT spacing, then at each end/hole you'll need the Force and Sense holes for the Kelvin Clip.

It's mostly going to be the same as V4 board pictured above. The V4 board has a 4mm SMT pad on top and bottom of the board running to each test point, with the bottom traces connecting via a hidden via near each component. The issue was only making the physical connection with the kelvin clips more reliable.

Instead, for the new version, I'll keep everything top side in parallel and eliminate the extra vias. As you can see from the test results, accuracy was great (at least for my needs). I'm not trying to get anywhere near metrology craziness. But considering the ST42 was direct to component, and the ST2832 was via the test points, that's certainly close enough for me. If any of the disagreements were related to the connection type, that would most likely be the fault of the inconsistency with the mechanical connection on the top/bottom pads that required solder to make the connection work.

I'll still be adding the open/short points as well for calibration before testing.

If you have any opinion about the above, please share. 😉

[KungFuJosh]

That leaves the question of the components.
Tight tolerance resistors are a dime a dozen, with capacitors it's a different story (2%), with coils the same (5%).

I have often wondered this, as I don't think I want to go to the effort of the large standard resistors but at the same time I would like some with good properties that mean consistency and little drift over time.

The resistors were the easiest and mostly the cheapest parts for the board. 0.1% at 10 to 15ppm for SMT resistors or 0.1% and 25ppm for the CMF industrial THT resistors. The rest of the board is trickier.

[Martin72]

@Josh:

1mm drill hole..

[mawyatt]

I always like to use these golden support points for my PCBs, they are suitable as test pins as well as wiring support points.
Narrow design but still very stable.
Edit: Pic added.

Your layout might not work with Kelvin Clips as they will interfere with each other near the handles (overlap), better to rotate the terminals 90 degrees so the Kelvin Clips can be on each opposite sides of the DUT facing away from the DUT.

Best