Active Probe Project

[smarley]

Hello everyone,

I’m working on an active probe project with the following specifications as a target:

Frequency Range: DC to 2 GHz
Attenuation: 10:1
Input Impedance: 1 MΩ // 1 pF
Output Impedance: 50 Ω

I’m currently finalizing the design and need some advice on the input and output connectors.

Output Connector:
Initially, I considered using an MMCX connector. However, MMCX connectors are typically rated for around 500 connections, which might not be sufficient for this application.
I’m now thinking of switching to an SMA connector for its durability and better connection reliability.

Input Connector:
For now, I’ve placed a standard header, but I’m searching for a better solution that allows for direct connection or adaptation for various needs.
Using an MMCX connector for the input was an option, but it would increase the input capacitance and has the same issue with the number of connections.
Directly using pogo pins is another possibility. However, they would be fixed in size, limiting flexibility and possibly not accommodating an adapter board.


Does anyone have a better solution for the tip connector that would provide flexibility and reliability?
Any recommendations on connectors that balance durability, low capacitance, and adaptability?

I appreciate any input or experiences you can share!

The project is inspired by: https://jmw.name/projects/active-probe/
Based on BUF802 from TI design (https://www.ti.com/product/BUF802).

I will share pictures of the PCB/Layout and housing soon.

Thank you!

[Postal2]

.... adaptation for various needs. ....

For what needs?

If you are making a probe with pin, then you should expect 200 MHz, 2 GHz is a different story.

[smarley]

.... adaptation for various needs. ....

For what needs?

If you are making a probe with pin, then you should expect 200 MHz, 2 GHz is a different story.

Yes, my goal is to get the 2Ghz with the short connection as possible, maybe with the header as they are today, I know that with the adapter this will be reduced.

About the adapters I was thinking to solve my daily problems, for example, an adapter with pogo pins on the distance to measurement 0603, 0805, 1206.
Or to connect some smd rf connector on the pcb for example.

[Postal2]

Yes, my goal is to get the 2Ghz ...

Okay. For a 2 GHz probe, the design must be fundamentally different. Again, why? Basically, at 2 GHz the signals are differential, but at 200 MHz they are not. How will you switch your probe?

I would be interested in a probe up to 100 MHz, especially for Fnirsi. Most likely, it should be powered by a small battery and not look like a heavy hammer.

In principle, your probe is good for me, if the price will not scare me away.

[jmw]

That front end layout looks quite familiar! Please tell me you are keeping the design open source in compliance with the CERN-OHL-S license

https://jmw.name/projects/active-probe/
https://github.com/jmwilson/oshw-active-probe

I went with SMA in my design because it's easy to buy and make cables, and adapters are plentiful.
Pins sockets are tricky but the Mill-max 0335 is compatible with PMK's probe tip accessories.

What's the USB-C connector for, and all the ICs and discrete transistors? Is it using USB-PD to charge the battery?

[smarley]

That front end layout looks quite familiar! Please tell me you are keeping the design open source in compliance with the CERN-OHL-S license

https://jmw.name/projects/active-probe/
https://github.com/jmwilson/oshw-active-probe

I went with SMA in my design because it's easy to buy and make cables, and adapters are plentiful.
Pins sockets are tricky but the Mill-max 0335 is compatible with PMK's probe tip accessories.

What's the USB-C connector for, and all the ICs and discrete transistors? Is it using USB-PD to charge the battery?

Hi, yes I saw your project before and it is very nice, congrats!
Let's see how will be my input impedance and S21, I can not use 0402 components otherwise I will have troubles to solder it, so we can compare the results later.
Also, I thought in different arrangements of the input front end that I want to test, my idea is do two or three different layouts to try.
I'm thinking if I will release as oshw also, I like the idea because then it could be easy for repair of the product in the future, and not depending on me to do it.

My project is based on BUF802 from TI design (https://www.ti.com/product/BUF802) also with other features.
The goal I have in mind is to achieve something similar with Tetris® 1500 from PMK.
What I'm most interested is in having a flat response up to 600-800Mhz.
Results will be shared here, I think I will be able to have it up to end of August.

I will take a look on the connector you suggested, which of PMK accessories it would be compatible with?

Thinking in my daily work, I decided to not rely exclusively on power supplies, so I added a battery to have fewer cables running around the setup.
Battery life is a topic of concern, I'm thinking to have an option to reduce the bandwidth to reduce the consumption in the future but it's not implemented at the moment.
USB-c connector is to charge the battery or feed the probe if you don't have time to charge it before.
The ICs and transistors you see are for charging the battery, boost converter, reverse polarity protection, low voltage protection, switch.

[smarley]

Yes, my goal is to get the 2Ghz ...

Okay. For a 2 GHz probe, the design must be fundamentally different. Again, why? Basically, at 2 GHz the signals are differential, but at 200 MHz they are not. How will you switch your probe?

I would be interested in a probe up to 100 MHz, especially for Fnirsi. Most likely, it should be powered by a small battery and not look like a heavy hammer.

In principle, your probe is good for me, if the price will not scare me away.

Yes, I will try to go as compact as possible, today is more os less 115mm x 25mm.
I will use 18650 battery on it.
Price I have no clue at the moment, but need to be accessible and not this crazy prices you see on Keysight active probes, etc...

[jmw]

Hi, yes I saw your project before and it is very nice, congrats!
Let's see how will be my input impedance and S21, I can not use 0402 components otherwise I will have troubles to solder it, so we can compare the results later.
Also, I thought in different arrangements of the input front end that I want to test, my idea is do two or three different layouts to try.
I'm thinking if I will release as oshw also, I like the idea because then it could be easy for repair of the product in the future, and not depending on me to do it.

My project is based on BUF802 from TI design (https://www.ti.com/product/BUF802) also with other features.
The goal I have in mind is to achieve something similar with Tetris® 1500 from PMK.
What I'm most interested is in having a flat response up to 600-800Mhz.
Results will be shared here, I think I will be able to have it up to end of August.

I will take a look on the connector you suggested, which of PMK accessories it would be compatible with?

Well you edited your response, but originally it said you did not copy/use the layout/schematic -- expect you definitely did, as evidenced by the layout is identical around the BUF802 and the TPS7A39. And those parts are certainly 0402; at this point I'm pretty sure of how they are proportioned in relation to those two ICs. Look, I open-sourced my design partly as an experiment to test the OSHW waters with something a little more substantial than a breakout board or LED badge. I want to continue to share OSHW projects, so if you want to incorporate the design into yours, please do so under the terms of the license. The responsibilities are described in CERN's guide: https://ohwr-gitlab.cern.ch/project/cernohl/-/wikis/uploads/cf37727497ca2b5295a7ab83a40fcf5a/cern_ohl_s_v2_user_guide.pdf

As for the tips, the PMK Tetris accessories share a pin diameter that is compatible with the 0335 socket. I've used the solid tip (890-800-000) and it fits the spring detent in the socket perfectly.

[smarley]

Hi, yes I saw your project before and it is very nice, congrats!
Let's see how will be my input impedance and S21, I can not use 0402 components otherwise I will have troubles to solder it, so we can compare the results later.
Also, I thought in different arrangements of the input front end that I want to test, my idea is do two or three different layouts to try.
I'm thinking if I will release as oshw also, I like the idea because then it could be easy for repair of the product in the future, and not depending on me to do it.

My project is based on BUF802 from TI design (https://www.ti.com/product/BUF802) also with other features.
The goal I have in mind is to achieve something similar with Tetris® 1500 from PMK.
What I'm most interested is in having a flat response up to 600-800Mhz.
Results will be shared here, I think I will be able to have it up to end of August.

I will take a look on the connector you suggested, which of PMK accessories it would be compatible with?

Well you edited your response, but originally it said you did not copy/use the layout/schematic -- expect you definitely did, as evidenced by the layout is identical around the BUF802 and the TPS7A39. And those parts are certainly 0402; at this point I'm pretty sure of how they are proportioned in relation to those two ICs. Look, I open-sourced my design partly as an experiment to test the OSHW waters with something a little more substantial than a breakout board or LED badge. I want to continue to share OSHW projects, so if you want to incorporate the design into yours, please do so under the terms of the license. The responsibilities are described in CERN's guide: https://ohwr-gitlab.cern.ch/project/cernohl/-/wikis/uploads/cf37727497ca2b5295a7ab83a40fcf5a/cern_ohl_s_v2_user_guide.pdf

As for the tips, the PMK Tetris accessories share a pin diameter that is compatible with the 0335 socket. I've used the solid tip (890-800-000) and it fits the spring detent in the socket perfectly.

Hi, sorry for the misunderstanding, yes you are right, I realized that after I post, it's the reason I edit it, this was the first version with 0402.
So indeed the picture is with 0402, and it will not work for me. I will update it later to have no mistakes here.
I was reading the CERN's license and I fully agree it, even if I don't do an OSHW I will respect it fully, specially the "3 How to deal with hardware designs licensed
under CERN-OHL-S v2" for sure.
I really appreciated you work and the approach with OSHW.

[smarley]

That front end layout looks quite familiar! Please tell me you are keeping the design open source in compliance with the CERN-OHL-S license

https://jmw.name/projects/active-probe/
https://github.com/jmwilson/oshw-active-probe

I went with SMA in my design because it's easy to buy and make cables, and adapters are plentiful.
Pins sockets are tricky but the Mill-max 0335 is compatible with PMK's probe tip accessories.

What's the USB-C connector for, and all the ICs and discrete transistors? Is it using USB-PD to charge the battery?

Hi James, added your webpage in the first post as a reference for the design.
Removed the old pictures and I will add the actual pictures of the layout/housing soon.

Thank you!