UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Test Gear and Workshop Equipment

Notices

Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Closed Thread
 
Thread Tools
Old 7th Dec 2020, 8:37 pm   #21
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: Scope probes: techniques and builds

typo, sorry.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 7th Dec 2020, 10:13 pm   #22
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Thumbs up Re: Scope probes: techniques and builds

O.K., David: thanks. For one minute, I thought I was about to learn something new . . . . again.

Al.
Skywave is offline  
Old 8th Dec 2020, 3:00 pm   #23
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Scope probes: techniques and builds

Quote:
I may be able to build my own probe suitable for a HP 8558B Spectrum Analyser; I am very interested in this idea and would love to give it a go.
Hi Bill, I looked up the specs for your analyser and it covers up to 1500MHz. The input VSWR with >= 10dB internal attenuation is better than 1.5:1 and the amplitude response vs frequency is about +/-1dB.

In reality I'd expect the worst case input VSWR (up towards 1500MHz?) to be closer to 1.2:1 with 10dB internal attenuation. This means that if you made a -20dB Zo probe using a series 450R resistor and some decent coax the overall +/- 1dB amplitude response would be degraded by up to about about +/- 0.6dB due to the mismatch uncertainty from the 1.2:1 input VSWR. In addition to this the self capacitance of the series 450R resistor could reduce the attenuation of the probe by 1-2dB up towards 1500MHz. So overall the accuracy of the probe could degrade a fair bit by 1500MHz. However, for use up to maybe 500MHz it would work very well with your analyser as long as you maintain at least 10dB internal attenuation in the analyser to maintain a low input VSWR at the analyser input port.

If you make the -40dB version of the Zo probe with the series 5k resistor then I think you can expect to see the attenuation of the probe drop quite a lot (10-15dB?) by 1500MHz due to the self capacitance of the series 5k resistor. A lot depends on what size/package the resistor is and it you choose to use several tiny resistors in series to make up the 5k total resistance. It is possible to compensate the 5k resistance with some shunt capacitance after the resistor and this would help to flatten the frequency response.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 8th Dec 2020, 9:30 pm   #24
Bufo Bill
Heptode
 
Bufo Bill's Avatar
 
Join Date: Jan 2016
Location: Wolverhampton, West Midlands, UK.
Posts: 777
Default Re: Scope probes: techniques and builds

Well this is progressing nicely, we have some designs I can try. Most of the science is basically new to me, so very exciting. I am having problems with components however. I am not sure how to select coax, do the numbers and letters signify some defining characteristics or are they simply catalogue numbers or some such?
Thanks to Jeremy for taking the time to post that detailed assessment, and to everyone who has posted.
Cheers from Bill.
Bufo Bill is offline  
Old 8th Dec 2020, 11:15 pm   #25
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: Scope probes: techniques and builds

RG means radio guide, a US military catalogue, now obsolete, but the numbers are cast in human memory.

I use the Huber and Suhner website to look them up, or Belden cables.

Beware though, these give the real data of the proper made-to spec cable. There is a lot of imported rubbish sporting those numbers as well. You need to know where it's come from to know what you 're getting.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 9th Dec 2020, 1:31 pm   #26
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Scope probes: techniques and builds

Yes, the cheapest and easiest way to find a cable is to use something like ebay. However, there will be some risk in terms of the quality as David says. The alternative is to buy it from a trusted retailer like Mouser or Farnell but the cost will ramp up a lot if you include delivery charges.

If you only want to probe up to a few hundred MHz then RG174 cable will be fine. This cable is thin and flexible and and this makes it easier to use and position the probe on a circuit. The cable outer is usually black for RG174.

If you want a bit more performance (especially up to the 1500MHz limit of your analyser) then I'd recommend KO2252/D cable. This cable is thin but not quite as wiggly/flexible as RG174. It's usually a light brown colour.

On ebay you can buy ready made RG316 BNC/SMA cables (up to about 1 metre long) for about £8. It's probably going to be OK for use with a Zo probe especially if you only plan to use it up to 300MHz or so.

If I get some free time later I'll try and show the difference between RG174 and KO2252/D cable when used in a x10 Zo probe up to 1500MHz.
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 9th Dec 2020 at 1:39 pm.
G0HZU_JMR is offline  
Old 9th Dec 2020, 1:58 pm   #27
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Scope probes: techniques and builds

I just looked up the specs for RG316 cable and there are two versions of RG316 cable. One has a single screen and one has a double screen (-DS version).

At work we use KO2252/D cable which is similar but double screened.

https://www.richardsonrfpd.com/docs/..._dataSheet.pdf


I suspect that the RG316 cables on ebay will be single screen rather than the RG316-DS cable type as in the link below.

https://www.pasternack.com/images/pr...F/RG316-DS.pdf

I'll do some comparisons later. I have some Suhner branded RG174 cable here and I think the comparison cables I have are made with KO2252/D (double shielded). I expect the KO2252/D to perform much better in terms of mismatch uncertainty although this will probably be the most significant up above 1GHz or so. If you mainly want to operate below 300MHz I'm not sure there will be much improvement over the cheapo RG174 alternative.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 9th Dec 2020, 3:40 pm   #28
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Scope probes: techniques and builds

I had a decent rummage in the garage and managed to find an old 25 metre reel of RG316/U cable. This has Farnell part 142-3322 stamped on it.

https://uk.farnell.com/carlisleit/mr...2?ost=142-3322

You can also buy this cable by the metre from Farnell but it costs about £6 a metre! Please don't rush and buy the RG316 cable from Farnell as it is very expensive.

It is single screened and slightly skinnier than the double screened K02252/D cable I have here. I expect the K02252/D to outperform it but I'll try it anyway. I also managed to find an old reel of unbranded RG174 cable with 'DO NOT USE!' written on it in bold marker pen. I suspect this unbranded cable will not work well as the Suhner branded RG174 cable up at 1GHz but it will probably be fine to use in a x10 Zo probe at 300MHz.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 9th Dec 2020, 9:32 pm   #29
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Scope probes: techniques and builds

I've done a few initial comparisons between cables on my VNA here at home. This is a fairly decent 4 port VNA and I've used port 1 and port 3 as these have the nicest spacing for my test fixture.

First of all I've plotted below the uncorrected response of a signal sent from port 1 to port 3 via a short (expensive!) Gore cable. This shows there is very little droop or ripple in the VNA even when it is uncorrected. So pretty much all of the ripple seen in the plots will be due to the cables.

I also used a pair of chip resistors (180R + 270R) to make up the 450 tip resistor. Finally, because I haven't done a full 2 port calibration of the VNA I've also added a precision 10dB attenuator (Narda 12GHz) to port 3 to make sure it has a good port match. The test fixture itself is a short run of microstrip on some Rogers PCB material and the return loss of the microstrip is better than 40dB up to 1500MHz. So this is a decent 50R test fixture.

The first plot shows the raw VNA response. The other plots are for the RG316/U cable from Farnell, the Suhner RG174 cable and the K02252/D cable when they are used in a x10 Zo probe with 450R tip resistance. The cables are not the same length as each other but they are similar lengths.

Note also that the VNA shows 30dB loss but this includes the 10dB attenuator so the Zo probe has just over 20dB loss as expected.

The double screened K02252/D cable is the clear winner although the other two cables perform OK up to about 300MHz. There doesn't seem to be much difference between the RG174 cable and the RG316/U cable. However, the nice thing about the RG316/U cable is that the dielectric is PTFE so it won't melt with a soldering iron. The dielectric inner of the RG174 is Polyethylene and this melts when heated. This makes it really easy to create an inner to outer short when soldering. So I would recommend the RG316/U over the RG174 unless your soldering skills are very good
Attached Thumbnails
Click image for larger version

Name:	S31raw.gif
Views:	69
Size:	18.3 KB
ID:	222224   Click image for larger version

Name:	Rg316u.gif
Views:	75
Size:	19.0 KB
ID:	222225   Click image for larger version

Name:	Suhner(rg174.gif
Views:	77
Size:	19.7 KB
ID:	222226   Click image for larger version

Name:	K02252(d.gif
Views:	71
Size:	18.5 KB
ID:	222227  
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 9th Dec 2020, 9:47 pm   #30
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Scope probes: techniques and builds

The reason there is so much ripple in the RG174 and RG316/U plots is because this is a very strict test of the accuracy of the characteristic Zo of the cable. The far end of the cable is going to be very close to a resistive 50R load because there is a precision attenuator followed by a 50R VNA port at the far end.

However, at the tip end of the cable this near perfect 50R resistance will be spoiled by the misbehaviour of the coax cable. If the VSWR looking into the coax cable at the tip end is 1.35:1 for example then there could be approx +/- 1dB of ripple seen in the probe response due to mismatch uncertainty between a 500R (ish) source via the tip resistor feeding a transmission line with a VSWR of 1.35:1.

This is why the quality of the cable and also the quality of the port VSWR of the analyser is so important in a Zo probe design. If sub standard parts are used then there will be a lot of ripple in the Zo probe response.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 9th Dec 2020, 11:15 pm   #31
Bufo Bill
Heptode
 
Bufo Bill's Avatar
 
Join Date: Jan 2016
Location: Wolverhampton, West Midlands, UK.
Posts: 777
Default Re: Scope probes: techniques and builds

Wow, thank you so much Jeremy, that's such a useful experiment. Thanks so much for taking the time to conduct and report your findings, it has been a big help to my practical understanding of the issues at play here.
Cheers from Bill.
Bufo Bill is offline  
Old 9th Dec 2020, 11:25 pm   #32
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Scope probes: techniques and builds

Hi Bill thanks for those comments! I was a bit disappointed by the performance of the RG316/U cable so I tried doing a time domain analysis of this cable on the VNA. This can show where there are mismatch problems along the cable length. The TD analysis showed that the worst of the mismatch is at the crimped RF connector at the far end of the probe cable. I fitted a cheapo crimped connector to the RG316/U cable and this may be why the overall VSWR of the cable is as poor as 1.35:1 at 1500MHz.

By contrast the K02252/D cable has a top quality SMA RF connector (professionally fitted!) at the analyser end and this cable looks much better on the VNA with the TD test.

I'll try and find a better quality connector for the RG316/U cable but it's still going to have to be some form of crimped SMA. I can try and find a genuine Suhner connector at work but I don't know when I'll be back working in the office.

The overall conclusion is still the same though. The quality of the cable is really important in a Zo probe and this includes the quality of the connector(s) at the end of the cable as the connector interface alone can spoil things.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 10th Dec 2020, 2:07 pm   #33
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Scope probes: techniques and builds

Over lunchtime I tried fitting a decent Suhner SMA connector on the end of the RG316/U cable and the results are now very good!

I don't have any Suhner (crimped) SMA connectors for RG316 but I do have some decent straight Suhner SMA connectors designed to be used with 0.086" diameter semi rigid cable up to about 18GHz. This is a bit of a dirty bodge as it requires the braid of the RG316/U cable to be heavily tinned with solder such that it appears as a solid jacket. The connector is soldered to the cable rather than crimped and the fitting procedure is attached and it is possible to fudge this type of connector to RG316/U cable. It wouldn't be recommended as a permanent solution for lots of reasons but the performance is now really good with this connector. See below. Much better! There is hardly any ripple at all now.

Note that this probe is still using 180R + 270R chip resistors in series to make up the 450R tip resistor. This is fairly impractical and not very rugged. If a conventional metal film resistor is used it will have higher self capacitance and the response will change. Instead of drooping about 1dB by 1GHz (due to the insertion loss of 1m of RG316/U at 1500MHz) the response will be lifted up by 1-2dB by 1500MHz and this depends on the self capacitance of the resistor.

The RG316/U cable used for the test below is about 112cm long.
Attached Thumbnails
Click image for larger version

Name:	Rg316u(sma.gif
Views:	87
Size:	17.3 KB
ID:	222251  
Attached Files
File Type: pdf HUBER+SUHNER_11_SMA-50-2-110_assemblyInstruction.pdf (194.2 KB, 62 views)
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 11th Dec 2020, 4:10 pm   #34
Bufo Bill
Heptode
 
Bufo Bill's Avatar
 
Join Date: Jan 2016
Location: Wolverhampton, West Midlands, UK.
Posts: 777
Default Re: Scope probes: techniques and builds

Thanks again Jeremy, hope you had fun doing this, it must have taken some time.
I will let you know how the builds go, I'm really looking forward to it!
Cheers from Bill.
Bufo Bill is offline  
Old 11th Dec 2020, 9:47 pm   #35
Bufo Bill
Heptode
 
Bufo Bill's Avatar
 
Join Date: Jan 2016
Location: Wolverhampton, West Midlands, UK.
Posts: 777
Default Re: Scope probes: techniques and builds

Okay, going back to basics, is there a way I can easily change the capacitance of a 1:1 Oscilloscope probe? For example, modern probes seem to be mostly 70pF. Is there simply a capacitor i could replace, or does the capacitance depend more on the connections of the various coax and probe tips? Not actually taken one apart before so it could be composed of shoelaces and banana skins for all I know!
Sorry if I'm being dim.
Cheers from Bill.
Bufo Bill is offline  
Old 14th Dec 2020, 12:36 pm   #36
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Scope probes: techniques and builds

Hi Bill
You can use a 1:1 scope probe with a spectrum analyser but it will load the circuit quite a bit. Usually, a 1:1 scope probe also has a 10:1 switch on it and the overall design of the probe is a compromise between these two settings. Rather than mess with the insides of a scope probe I think it would be better to make something using regular coax cable like RG316 or RG174.

The coax cable used in a 1:1 / 10:1 scope probe will be deliberately lossy and often the centre conductor of the coax will be made with a form of resistance wire. If you connect this to a spectrum analyser with a 50R input then this isn't what the probe was designed for. it will still act as a probe but in 1:1 mode the bandwidth will be limited to maybe 20MHz and the input impedance of the probe will be low. It will be lower than the x10 Zo probes already mentioned on the thread. If you switch the 1:1 probe across to 10:1 and connect it up to a 50R spectrum analyser then I think the probe will show a high pass filter response. It will be very deaf across the LW/MW bands for example.

You can definitely still do useful work with a 1:1 or 10:1 scope probe connected to a spectrum analyser but it is far from an ideal setup.

One other option (if you have a Tek 465 scope) is to connect the Channel 1 output (BNC) connection at the rear of the scope to the 50R input of a spectrum analyser. Then fit a regular 10:1 scope probe to the channel 1 input on the front panel of the scope. This way you probe a circuit with the 10:1 scope probe and you can see the response in the time domain on the scope CRT but you can also see the frequency response on the spectrum analyser via the rear panel CH1 OUT connection of the scope. The setup will be limited by the 100MHz bandwidth of the scope and also the 10:1 scope probe will load the circuit quite a bit for signals above a few MHz. However, for IF and AF work the above setup can work quite well. A frequency counter can also be used in place of the spectrum analyser and that is how I used to configure my Tek 465 and frequency counter when I used to tinker with CB/Ham radio stuff in my youth.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 14th Dec 2020, 12:50 pm   #37
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: Scope probes: techniques and builds

The gentle art of knowing what you can get away with! It has to be matched by knowing what is pushing it too far.

I use a 10:1 scope probe into a 50 Ohm analyser quite often for a quick shufti to get an idea whats going on. I can have no trust in amplitude calibration or flatness, but I can compare adjacent signals, and I can look for things that shouldn't be there. Measuring their levels is a lot more formal, but the quickie tells me whether or not to bother.

I was going to say cheating is a sign of intelligence, but the world stage disproving that just at the moment. Let's say knowing how and when to cheat can be very useful (and when to stop)

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 14th Dec 2020, 1:48 pm   #38
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Scope probes: techniques and builds

Agreed! I've used a scope probe with a spectrum analyser many times and in some cases the task/solution can be achieved just by holding the tip of the probe fairly close to the test point. As it isn't touching anything in the circuit this doesn't load the circuit at all and the (sensitive) analyser will still be able to display something useful in the frequency domain even though the absolute level isn't calibrated in any way.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 14th Dec 2020, 11:31 pm   #39
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Scope probes: techniques and builds

Quote:
Originally Posted by woodchips View Post
I would be surprised if a 300MHz scope was only flat to 100MHz. The bandwidth is normally specified at the front panel connector, so will be 300MHz -3dBm.
I'm not disputing the -3 dB freq. at 300 MHz. It's the 'reasonably flat' freq. that is important here (for me) for making reasonably accurate measurements with a 'scope at relatively high frequencies. (And even that ignores the B/W of any probe that may be in use ).

I've made a measurement of the B/W of my Tek. 2465 300 MHz 'scope.
For a source, I used an HP 8640B signal generator at 10 MHz with 10 mV rms output (to 50 ohms), fed to the 'scope with a 1m. length of RG55/U coax.

A preliminary check of this measurement system was performed first. The sig. gen. was connected to an Advantest spectrum analyzer (50 ohm input) with that length of coax. to check that the freq. response of the measurement system was 'flat' to > 300 MHz: it was, and that the chosen output level of the sig. gen was correct: 10 mV rms at 10 MHz to > 300 MHz: it was.

The Tek. 2465 has two main input channels: I tested only one. Each input can be selected for a.c. hi-Z input or internal 50 ohms (amongst others). I made measurements for both input configurations.

Results.
1. Internal 50 ohm selected.
(a) -3dB point.
Input: 10 mV rms = 28 mV p-p.
-3 dB on 28 mV p/p = 19.8 mV p/p
For 19.8 mV p/p on 'scope screen, input freq. = 300 MHz (approx.)

(b) -1 dB point.
-1 dB on 28 mV p/p = 25 mV p/p.
For 25 mV p/p on 'scope screen, input freq. = 135 MHz (approx.)

(c) -0.5 dB point.
-0.5 dB on 28 mV p/p = 26.4 mV p/p.
For 26.4 mV p/p on 'scope screen, input freq. = 120 MHz (approx.)

2. External 50 ohm selected.
Same measurements and measurement conditions as above.
Results.
-3 dB @ 260 MHz; - 1dB @ 135MHz; - 0.5 dB @ 125 MHz.

(Differences between measurement results 1 and 2 are put down to inherent measurement errors - and are not considered to be significant).

Notes.
1. The Tek. 2465 has the facility to measure the peak-peak voltage displayed on its CRT and to display that measurement as numbers and units of measurement on the CRT. That is done by a pair of movable horizontal cursors: a very convenient way of measuring signal amplitude.
2. The external 50 ohm load used for test 2 was an HP lab. standard coaxial 50 ohm load. I keep this item reserved for exclusive use with low-power calibration checks.
3. I did not repeat all of the above measurements at different signal input levels. The measurements were intended only to ratify the claimed -3 dB B/W freq. and to determine - albeit approximately - the 'flat' freq. response. The determination of the linear dynamic range of the 'scope was not intended nor performed: that is a different task.

Final comments.
1. Note that the ratio 'flat freq'. to the -3db freq. is in the region of 0.4 (120 MHz ÷ 300 MHz).
2. In the past I have owned various Tek. and HP 'scopes (with claimed -3 dB B/Ws of 100 MHz or more) and have found that that ratio is quite typical.
3. All of the above text took about 20 times longer to write and proof-read than it did to actually perform the said measurements!
I hope others here will find the above useful.

Al.
Skywave is offline  
Old 15th Dec 2020, 12:04 am   #40
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Default Re: Scope probes: techniques and builds

Re: post 38: using a 'scope probe with a spectrum analyzer to 'investigate' R.F circuitry.
A few years ago when I was employed in the VHF / UHF radio transmission equipment design, manufacture and service world, I frequently used that approach to search for unwanted harmonics & spurii emanating from high-power R.F power amplifiers. Apart from keeping the probe not too close to power filters, Wilkinson splitters, hybrid couplers, etc., being careless occasionally caused a very unpleasant R.F. burn!

Al.
Skywave is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 2:58 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.