50 Ohm Instrument Probes - A Possible New Product

[German_EE]

Scope probes are designed to work with oscilloscopes, that's obvious, however they can also be used with a frequency counter when moving around a circuit. Scope probes however fail when it comes to things like spectrum analysers because a scope probe is designed for a 1M ohm environment but a spectrum analyser expects to see 50 ohms.

So, why not have a probe that connects the probe tip to the coax inner and a flying lead to the coax shield, with a BNC plug at the other end of some 50 ohm coax? Over the weekend I contacted a local manufacturer and their reply was less than enthusiastic, they were unwilling to commit to even a small production run because they had no idea how many they would sell. They have a valid point, nobody wants to waste money these days.

How many of you would be interested in purchasing one or more of these probes? Please note that I am NOT selling these and I am NOT looking for orders, all I want is a reply to this post saying how many 50 ohm probes you would purchase, either for yourself or the company that you work for. If there is enough interest then I will pass the numbers (not your details, just the numbers) to the probe manufacturer and we can get an initial production run.

[EEVblog]

50ohm high bandwidth coax test cables are a classic DIY design. Plenty of info in IIRC, a High Speed Digital Design a Handbook Of Black Magic and other texts.
http://www.sigcon.com/Pubs/straight/probes.htm

[ogden]

So, why not have a probe that connects the probe tip to the coax inner and a flying lead to the coax shield, with a BNC plug at the other end of some 50 ohm coax?

Such 50ohm probe is useless. If you connect 50ohm probe in parallel to 50ohm circuit, impedance will be no more 50hms, signal amplitude will go down and due to load/impedance mismatch your circuit may not operate as supposed just because you are probing. For RF probing you need passive attenuator probe or active one, like this:

http://www.instructables.com/id/DIY-1GHz-Active-Probe-for-Under-20/

[edit] Most RF engineers does not usually probe but test parts of RF circuit (modules) individually - by taking them out of signal chain.

[tggzzz]

Scope probes are designed to work with oscilloscopes, that's obvious, however they can also be used with a frequency counter when moving around a circuit. Scope probes however fail when it comes to things like spectrum analysers because a scope probe is designed for a 1M ohm environment but a spectrum analyser expects to see 50 ohms.

So, why not have a probe that connects the probe tip to the coax inner and a flying lead to the coax shield, with a BNC plug at the other end of some 50 ohm coax? Over the weekend I contacted a local manufacturer and their reply was less than enthusiastic, they were unwilling to commit to even a small production run because they had no idea how many they would sell. They have a valid point, nobody wants to waste money these days.

How many of you would be interested in purchasing one or more of these probes? Please note that I am NOT selling these and I am NOT looking for orders, all I want is a reply to this post saying how many 50 ohm probes you would purchase, either for yourself or the company that you work for. If there is enough interest then I will pass the numbers (not your details, just the numbers) to the probe manufacturer and we can get an initial production run.

Search for "resistive divider" or Z0 probes.

The flying ground lead will have 1nH/mm of length. Consider how that will interact with any capacitances near the tip.

[T3sl4co1l]

Just get a preamp to use your 1M probe on a 50 ohm system.

Bonus points: obtain a mate for your favorite instruments' "PROBE POWER" connector and use that instead of an external adapter.

Tim

[rfeecs]

50 ohm RF probes have been around for decades.  Usually used for wafer probing.  But can also be used for PC board probing:

https://www.cascademicrotech.com/products/probes

http://www.ggb.com/

[ogden]

But can also be used for PC board probing:

Wrong. Probes for automated RF component testing have very specific coplanar waveguide tips, usually manufactured according to chip dimensions/specs. Such definitely cannot be used for PC board probing.



Anyway buying such very expensive, application specific probe just for PC board "poking" would be quite dumb idea.

[rfeecs]

But can also be used for PC board probing:

Wrong. Probes for automated RF component testing have very specific coplanar waveguide tips, usually manufactured according to chip dimensions/specs. Such definitely cannot be used for PC board probing.

Typically they are not custom made.  They come in standard pitches.  They do require a probe pattern on the chip to match the probe pitch.

Obviously for a PCB, you would not use the same probe that is used for on wafer probing.  Here is an example of pcb testing:
http://www.ccnlabs.com/probestations.html

Anyway buying such very expensive, application specific probe just for PC board "poking" would be quite dumb idea.

Definitely.  But it is pretty much an example of what German_EE has described.  50 ohms from end to end.

[ogden]

Obviously for a PCB, you would not use the same probe that is used for on wafer probing.  Here is an example of pcb testing:

Agreed. 50ohm probes indeed exist I doubt that OP wanted to manufacture hi-end probes for RF manufacturing anyway. BTW this one looks like good RF probe for PCB probing, price hard to beat:

https://www.ebay.com/itm/RF-Active-Probe-0-1-1500-MHz-1-5-GHz-analyzer-oscilloscope-RF-cable-included-/332393676082?hash=item4d6434fd32

[CNe7532294]

Just get a preamp to use your 1M probe on a 50 ohm system.

Bonus points: obtain a mate for your favorite instruments' "PROBE POWER" connector and use that instead of an external adapter.

Tim

https://www.ebay.com/itm/High-Impedance-Buffer-Amplifier-1MOhm-FETAMP1-Similar-to-Tektronix-TCA-1MEG/260750545356?hash=item3cb5f191cc:g:rLAAAMXQdohREzhS

Obviously for a PCB, you would not use the same probe that is used for on wafer probing.  Here is an example of pcb testing:

Agreed. 50ohm probes indeed exist I doubt that OP wanted to manufacture hi-end probes for RF manufacturing anyway. BTW this one looks like good RF probe for PCB probing, price hard to beat:

https://www.ebay.com/itm/RF-Active-Probe-0-1-1500-MHz-1-5-GHz-analyzer-oscilloscope-RF-cable-included-/332393676082?hash=item4d6434fd32

From a sales point of view see both ebay links click on "sold". Been getting cheaper options with more buys thru recent years if this is what OP was asking for. It all depends exactly on what specs you're providing us with.

[HighVoltage]

Philips offered 50 Ohm oscilloscope (LowZ) probes already 30 years ago.
I think I still have one of them, that came with the 200 MHz PM3394B Combiscopes.

 

[ogden]

Philips offered 50 Ohm oscilloscope (LowZ) probes already 30 years ago.
I think I still have one of them, that came with the 200 MHz PM3394B Combiscopes.

HP/Agilent/Keysight *had* 50ohm probes as well, such as 10437A and 10442B. Both obsolete now. Main question here is - who would need one *today*. I don't. Apart from SMA cables I use most, I have DIY divider "probe" like this:

https://www.embedded.com/electronics-blogs/break-points/4376054/A-cheap--high-quality-probe

Apart from already mentioned solutions, RF engineers may use RF test sockets/connectors like this:
https://www.murata.com/en-us/products/connector/switchconnector

[calexanian]

Anybody who works in wireless anything needs one now. Nearly all wireless devices are a 50 ohm load. Using anything other than a 50 ohm probe at high frequency will have an errant reading or distorted waveform and think they have harmonics when in reality its just a bad match. Any old ham will tell you that.

[ogden]

Anybody who works in wireless anything needs one now. Nearly all wireless devices are a 50 ohm load. Using anything other than a 50 ohm probe at high frequency will have an errant reading or distorted waveform and think they have harmonics when in reality its just a bad match. Any old ham will tell you that.

Ues, Mr.Obvious. You are right. 50 Ohms it is.

I have to repeat: if you *probe* with 50ohm probe, it's impedance will add to impedance of the circuit, resulting in load mismatch. Obviously it can disrupt operation of circuit, usually it does. So to probe internals of some circuit with 50 Ohm probe, you have to disconnect load of the circuit so it is loaded just with 50ohms of the probe. Then why you need probe? - Use just 50 ohm cable!

This is how most RF engineers work BTW - test modules individually by disconnecting downstream signal chain to connect spectrum analyzer, VNA, power meter or other RF equipment directly to output of module under test. They usually do not need 50ohm probes. Either hi-impedance RF fet preamp or just 50 Ohm cable with or w/o connector in the end.

[joeqsmith]

I have attempted to make my own passive probes for digital work and have had some decent results up to about 1.5GHz.   Normally the DC loading is around 1K.   Plot showing one of them using some 1206 parts. 

I planned to make a buffer as previously suggested and made it as far as playing with various op-amps on some protoboards.  The problem is they worked so well, I never took the next step to finalize it.   This does work well in some cases.   This is pretty limited bandwidth.  Normally when I am using this setup, its below 50MHz. 

For the higher speeds, higher impedance, I use active probes. 

This link has a picture showing some of my attempts at home made probes.
https://www.eevblog.com/forum/testgear/show-us-your-probes!/

[VK5RC]

A brief comment is that at any decent RF, say above 100MHz to 'probe' you often don't need to touch, but even coupling will potentially adversely effect your circuit function, so bugger up what you are trying to find out.

[ogden]

I have attempted to make my own passive probes for digital work and have had some decent results up to about 1.5GHz.   Normally the DC loading is around 1K.   Plot showing one of them using some 1206 parts. 

Now we are talking. Probe with impedance measurement is much more dependable that one w/o I made mine out of ~5cm semirigid microwave cable /w SMA in the end, 0603 size resistor and crimped tip of the small RF connector, everything heatshinked for rigidity. Results were more or less like yours, 1:10 at 900MHz. Luckily had to use just once.

I really like your 60KV scope probe. Serious stuff 

[Alex Eisenhut]

Tektronix Type 282

http://w140.com/tekwiki/wiki/282

Tektronix P6032

http://w140.com/tekwiki/wiki/P6032

[Insatman]

Just get a preamp to use your 1M probe on a 50 ohm system.

Bonus points: obtain a mate for your favorite instruments' "PROBE POWER" connector and use that instead of an external adapter.

Tim

Any suggestions for such a preamp?   

This one looks great but $400 plus shipping!
 https://www.ebay.com/itm/High-Impedance-Buffer-Amplifier-1MOhm-FETAMP1-Similar-to-Tektronix-TCA-1MEG/260750545356?hash=item3cb5f191cc:g:rLAAAMXQdohREzhS

[Alex Eisenhut]

Any suggestions for such a preamp?   

https://www.ebay.ca/itm/TEKTRONIX-015-0074-00-TYPE-282-PROBE-ADAPTER/222286927488?hash=item33c1555680:g:wqwAAOSwmLlYB5Pj

Price is decent too.

[joeqsmith]

I really like your 60KV scope probe. Serious stuff 

Thanks.  That was a fun problem to work out.   This is the original toilet paper roll prototype.