Scope Current Probe Deskew Fixture PCB Project

[Ribster]

_Wim_ and Ribster, would you be able to post schematics of your projects?

Steve,

I've added the schematic to the design.
It's not all that complicated. I've reused your NE555 circuit with some fixed resistors. I'm just opting for a fixed frequency and fastest rise time possible.
The project is done in Altium Designer. If you need the project, PM me.

I don't know if you can see it or not, but i via stiched the whole ground plane. Look at the attached schematic.
If you look at page 3 of my pdf, you see the top copper in red, and all those green dots. Those are stitching via's.
So i think my ground inductance will be okay, since the bottom copper layer is almost uninterrupted.
I could add some more capacitance on the other side if required. I could experiment with that once i get them built up.

I would suggest looking at the BSS316N MOSFET. This is in a more friendly SOT-23 package.

If I compare gate charge:

BSS316N

Well that's a nice find.
I've added the part to my parts library. I'll order some samples of them together with my parts order for this.
It should be straight forward to replace the GaN mosfet with this one, and compare!

@All:
What do you think about the solution proposed with the molex connectors ?
Those are plugin connectors, so you can add a wire or a coil, but its unpluggable, so it's handy to handle.
I'm just a bit worried about the distance between the connectors (net distance), but you can have the loop coming forward.
I don't see the use of making the pcb bigger for that reason. And i quite like the odd shape.

This is the male PCB side
https://www.digikey.be/product-detail/en/0395026002/WM25530-ND/2735307?utm_campaign=buynow&WT.z_cid=ref_octopart_dkc_buynow&utm_medium=aggregator&curr=eur&site=us&utm_source=octopart

This is the female plug in side
https://www.digikey.be/product-detail/en/molex/0395005002/WM23176-ND/2735117

[_Wim_]

Do not oversize the MOSFET this is will slow it down.

You are looking for low gate charge.

The RDS(on) could be 1 .

I personally would avoid an RDS(on) of 1 ohm. This means about 20 mA keeps flowing through the coil during the "off" period (with the resistances I used), and as some current probes have it difficult to nicely return to zero after a steep peak, I find it important that the off period is as close to zero current as possible. This then also tells you something about your probe. I see this more as a "current probe test fixture" than as a "deskew-only fixture".

So I would advocate for a low RDS on also, but I agree also to take a small FET with low input capacitance, although a little higher input capacitance can be covered by putting the clock driver really close.

BTW, I started first without the clock driver, but got rise time similar to the results from steve1515.

[_Wim_]

@All:
What do you think about the solution proposed with the molex connectors ?
Those are plugin connectors, so you can add a wire or a coil, but its unpluggable, so it's handy to handle.
I'm just a bit worried about the distance between the connectors (net distance), but you can have the loop coming forward.
I don't see the use of making the pcb bigger for that reason. And i quite like the odd shape.

What is the remaining net distance? This we could compare to the width of some of the "typical" probes. I will start with mine:

- Fluke i50A:  17mm
- Pearson 411: 26 mm
- Singer CP315: 32 mm
- ILA SMZ802: 27 mm

(yes, I had a TEA-current-probe-problem a while ago  )

But anyway, your current pcb footprint allows for the use screw terminals also, so I think the design is flexible.

[Elasia]

_Wim_ and Ribster, would you be able to post schematics of your projects?

Steve,

I've added the schematic to the design.
It's not all that complicated. I've reused your NE555 circuit with some fixed resistors. I'm just opting for a fixed frequency and fastest rise time possible.
The project is done in Altium Designer. If you need the project, PM me.

That looks pretty good, i'll have to remember to hit you up once its done being hashed on here less you eventually attach the ad files

[Ribster]

So i've gotten around to assembling them.
I need to still solder on the mosfets. Then i'll test the BSS316N and the GaN mosfet.

Will report back when testing is done

[steve1515]

Here's my board all finished up. 
What do you guys think?


Here are some of the measured rise times. The first image is with 1 loop and the second image is 8 loops.






I noticed if I measure at the input to the loop, I get a voltage spike. This is most noticeable with the 8 turn loop. Can someone check the Siglent board to see if the same thing happens?

[Jay_Diddy_B]

Hi,

In the LTspice simulation that I showed a few messages ago, you get the voltage spike if you connect the voltage probe to the top end of the coil.

Try connecting the voltage probe to the other end of the coil.

Regards,
Jay_Diddy_B

[theatrus]

I was just looking for a de-skew board since I'm not paying $700 for a PCB with a 555 timer - this design looks nice.

[Thebrakshow]

@Ribster
Did you ever get a chance to test out your fixture? I've been really interested in making a fixture for my Tek TCP312 and lecroy CP030 probes. Your design looks really nice. If you haven't had a chance to test it and are willing to open up the design, I for sure will be willing to give it a go!

Thanks!
Brak

[Ribster]

Hey there!

My design works and i’ve made the prototypes.
A forum member here contacted me and i sold him my prototype. If you’re interested i’ll produce a batch of 20 more and sell them on a tindie store.

Is there interest in buying a unit?
You can always hit me up at shop (at) ashlabs.be
I’ll send some scope plots or the unit with the GaN mosfet. Which i will produce 20 of.

Kind regards

[Thebrakshow]

Thanks for the offer, however, I was planning on making my own with some tweaks (to accommodate parts I already have in my lab). I was more so curious about the performance. Mainly curious if you ultimately went with the GaN (EPC2014C) or mosfet (BSS316N) or were both acceptable? What kind of rise times were you getting? When you use the fixture are you happy with the 555 timer as the clock source at ~7.6KHz? Or do you find yourself using an external clock source?

Thanks for your input!
As always you guys (entire EEVBLOG community ) rock 🤟!

-Brak

[blurpy]

@Ribster
Where do you hook up the regular oscilloscope probe on your design? Or is yours not made for the same purpose as steve1515s design?

[Thebrakshow]

I finally got around to building one of these fixtures. I just recreated Ribsters design / layout. The main modification I made was that I added footprints for both BSS316N and EPC2014C FETS. I've only tested out the BSS316N so far with pretty good results.

@blurpy I know this is an old post but if you look at Ribster's design it, has test point locations which you can use to probe the voltage with the tip and barrel method. I've attached a photo that should clear it up.

[blurpy]

@blurpy I know this is an old post but if you look at Ribster's design it, has test point locations which you can use to probe the voltage with the tip and barrel method. I've attached a photo that should clear it up.

Thank you, that explains it

[Ribster]

@Ribster
Where do you hook up the regular oscilloscope probe on your design? Or is yours not made for the same purpose as steve1515s design?

I've added 2 points that can be soldered on.
So you can attach it there.

I've finished the product including 3d printed case.
Only 1 is left at the shop, but ordered more PCB's for a new batch.
So if you want one, head over there