UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio 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.

Reply
 
Thread Tools
Old 11th Feb 2019, 6:03 pm   #21
GMB
Nonode
 
GMB's Avatar
 
Join Date: Aug 2003
Location: near Reading (and sometimes Torquay)
Posts: 2,179
Default Re: Looking to build RF detector probe for low frequency

Do bear in mind that RF doesn't like to go through metal. People often forget this. The earth current will try to go as close as it can to the signal current.

So a connection on the side will flow round the outside and go in through the hole where the signal wire enters. It then will try to follow the signal wire as close as it can get. It is all about minimising the magnetic field that the loop creates.

So don't connect your earth jack miles away from the signal input!
GMB is offline   Reply With Quote
Old 12th Feb 2019, 2:10 am   #22
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,545
Default Re: Looking to build RF detector probe for low frequency

The Haig probe circuit/design looks OK to me but it does have a few design limitations that could cause some confusion in certain circumstances. I hope it's OK to list these issues...

When it is in low gain mode I think that the J310 JFET will generate a significant amount of negative resistance at its gate. This would rarely cause problems but in theory at least it could become an oscillator when probing certain (inductive) loads. To null the negative resistance something like a series 82R resistor could be fitted inline with the 220pF cap at the input. This will degrade the input impedance a bit in both high and low gain modes but not by anything significant. I suppose that if you didn't bother with the 82R resistor it might be possible to exploit the negative resistance because you could use an inductor as a built in test load. If the probe is working it would oscillate (in low gain mode) with the required amount of inductance across the input.

So if the JFET and the diodes were OK then there would be a decent DC level seen at the detector output as it would detect its own self oscillation. So you wouldn't need an RF signal source to test the health of the whole RF probe/detector!

When it is in high gain mode, the input impedance will drop quite a bit by 4MHz. Much worse than a x10 scope probe in this respect. It could easily look like 15k ohm in parallel with 20pF by 4MHz. So it could cause significant circuit loading or pulling. By 10.7MHz the probe could look like 3k ohm in parallel with 20pF.

In high gain mode I'd expect the detector response to start rolling off by 5MHz. It could easily be -3dB by 20MHz. Not a problem if the upper limit of use is only 3 or 4MHz but it may cause disappointment to users hoping to try it out across the whole HF band up to 30MHz for example.

It's only going to have limited signal handling capability.

I'd have thought that a classic diode based RF probe would be more versatile than the Haig probe. Vastly improved bandwidth, lower circuit pulling/loading in many cases, cheaper to make, doesn't need a battery, easier to hold/handle and it can handle/detect much larger RF signals and it will easily work down to 50kHz.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Old 12th Feb 2019, 2:24 am   #23
Bazz4CQJ
Nonode
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 2,193
Default Re: Looking to build RF detector probe for low frequency

Possibly of interest, there is a site on which some people opened up some RF probes made by Boonton. The major objective was to try and see what could be leaned about the diodes, but the general design idea is worth noting.

For some reason, I cannot make the direct link work, but if you put

members.shaw.ca/novotill/RfProbeSensorMvUw/index.htm

in to Google, that will find three hits that take you there.

B
__________________
We know there are known unknowns. But there are unknown unknowns – ones we don't know we don't know.

Last edited by Bazz4CQJ; 12th Feb 2019 at 2:32 am.
Bazz4CQJ is offline   Reply With Quote
Old 12th Feb 2019, 2:43 am   #24
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,545
Default Re: Looking to build RF detector probe for low frequency

One thing to note with these RF probes it that I don't think it's a good idea (in terms of health and safety) to put them in an uncoated metal tube or box if they are being used with a handheld DMM.

If the operator is touching the metal case of the probe in one hand and is touching earth with another part of their body then they risk getting a nasty shock (or an RF burn?) if the probe croc lead can fall off and touch a live node on the equipment.

In all my years, I've never bothered with a screened RF detector probe and it doesn't seem to matter that much. My RF probe is based on the same cheap and throw away design I used as a student. It's just a long and thin strip of FR4 PCB cut in the shape of a pen. The handle section is just bare FR4 and only the very tip of the probe has any copper or detector components on it. This way the operator keeps their fingers away from any electrical connections (including the earth screen) when using the probe.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Old 12th Feb 2019, 4:54 am   #25
IsquaredR
Triode
 
Join Date: Apr 2017
Location: Milwaukee, Wisconsin, USA.
Posts: 34
Default Re: Looking to build RF detector probe for low frequency

Thanks G0HZU for your input. Perhaps it should be mentioned here that the metal box is connected to no part of the circuit. The probe, earth ground, "B- side" of the circuit and output to an oscilloscope have all been wired to separately from the metal box. Do you have trepidation about doing it that way? GMB my earth ground is about 8-inches long so I am hoping to obviate any circulatory current around the box.
This evening I worked just on the ground clip portion and have not had a chance to do any real testing which hopefully I will tomorrow. I am anxious to see how it does with an actual IF waveform.
IsquaredR is offline   Reply With Quote
Old 12th Feb 2019, 4:59 am   #26
IsquaredR
Triode
 
Join Date: Apr 2017
Location: Milwaukee, Wisconsin, USA.
Posts: 34
Default Re: Looking to build RF detector probe for low frequency

As a follow-up to a previous post from StuartH I did end up going with the ground clip being detachable. (See photo). As you can see the lead length is very short.
Attached Thumbnails
Click image for larger version

Name:	Haigprobe_4.jpg
Views:	42
Size:	87.6 KB
ID:	178246  
IsquaredR is offline   Reply With Quote
Old 12th Feb 2019, 10:27 pm   #27
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,545
Default Re: Looking to build RF detector probe for low frequency

Hi IIR,
For the health and safety comments I was really referring to the classic passive RF probes in the link provided by Bazz4CQJ. This classic screened design is for use with a (grounded?) VTVM but it often gets copied and used with a (floating?) handheld DVM instead.

The design of the floating probes is often based on old designs like the one below where the body of the probe is all metal. Modern designs often use a bare metal tube and the probe croc is soldered or screwed to the tube. In my opinion this type of probe + DVM would be risky to use on vintage tube gear because of the risk of a shock if the croc lead falls off and touches a high voltage point.

It's easy to see examples of probes like this online with a google search for:

RF probe 1N34A

Many of them are built inside an uncoated metal tube and used with a handheld DVM and this doesn't seem like a good idea unless the probe is only used on low voltage, low power equipment.
Attached Thumbnails
Click image for larger version

Name:	RF Probe for VTVM.jpg
Views:	49
Size:	71.3 KB
ID:	178289  
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Old 17th Feb 2019, 4:24 am   #28
IsquaredR
Triode
 
Join Date: Apr 2017
Location: Milwaukee, Wisconsin, USA.
Posts: 34
Default Re: Looking to build RF detector probe for low frequency

Quote:
Originally Posted by IsquaredR View Post
. . . and will certainly update this thread with my results and will include photos of my finished probe.
I've had ample time to familiarize myself with the Haig RF detector probe. It does a nice job detecting modulation/waveforms from an RF signal and is amply sensitive at low IF levels. My only complaint is that it does indeed have a tendency to skew the oscilloscope display with my hand capacity. Perhaps I did make a mistake by using a metal box even though its metal was isolated from the detector circuit. Now I am toying with the idea of removing the probe tip, replacing it with a BNC connector and using a shielded normal 10X oscilloscope probe. The detector could then lay at rest on the bench without hand capacity in proximity to it. I am also considering building a second Haig but, this time, installing it in a plastic box instead of a metal box.
IsquaredR is offline   Reply With Quote
Old 17th Feb 2019, 5:20 am   #29
Argus25
Octode
 
Argus25's Avatar
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 1,912
Default Re: Looking to build RF detector probe for low frequency

Its quite practical and possible to have a passive diode probe that is flat from low to very high frequencies. For example the HP probe type 11096B (schematic attached) It has a frequency response from 10kHz to 500MHz. Its voltage range is 0.2V to 30V rms and 200V DC max. It is designed to be loaded into a 10M Ohm meter load. It responds to the peak input value but is calibrated to read in rms.

Its input impedance is 4M Ohm shunted by 3pF max.

Simple circuit attached.

It has the usual input capacitor and rectifier CR1, however the resistive voltage divider consisting of R1,R3 in series with CR2 divides the peak voltage down to rms values.The trick is, at low input voltage levels, the non-linearity of the detector diode CR1, is compensated for by CR2. On the linear portion of the diode detector curve the division ratio of the divider CR2,R2 and the 10M load remain constant and it gives the 0.707 of peak. So CR2 compensates for the non-linearity or CR1 by proportionally altering the division ratio of the divider.

....clever, isn't it ?

The way I have approached the problem in the past is a use flat response wideband low capacitance input amps to get the voltage over a volt and then used precision detectors or diode detectors operating in their linear region.
Attached Thumbnails
Click image for larger version

Name:	rfprobe.jpg
Views:	54
Size:	59.9 KB
ID:	178542  
Argus25 is offline   Reply With Quote
Old 17th Feb 2019, 10:31 am   #30
MrBungle
Nonode
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 2,761
Default Re: Looking to build RF detector probe for low frequency

That is a bit clever. I’ve got itchy soldering fingers now

Edit: I suspect that would work well with the 1n5711 (poor mans HP diodes )
MrBungle is offline   Reply With Quote
Old 17th Feb 2019, 11:49 am   #31
Argus25
Octode
 
Argus25's Avatar
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 1,912
Default Re: Looking to build RF detector probe for low frequency

There are a couple of other methods that are equally clever.

As noted by H&H in their textbook, one way to overcome the non-linearity of a diode is to drive it with a current rather than a voltage source. In this instance, the signal of interest is converted to a current by a simple transistor circuit. This drives the diode rectifiers. What happens is the actual diode drive voltage rises to whatever value is required to program the correct diode current proportional to the input voltage through the diodes. I have posted this before on another thread, but I think the whole notion was under appreciated, because the circuit just looked like some sort of amplifier driving the diodes, but it was much more sophisticated than that. Their circuit can be improved a tad by adding a Fet at its input, to reduce any loading of the circuit under test.

Also, if you are dealing with very low level RF signals, as often is the case in Radio Astronomy, there is another very interesting trick. If a Tunnel diode is used in reverse (called a back diode) it acts like a diode with zero forward voltage drop, but it only works for low level signals because its effective PIV rating is quite low.
Argus25 is offline   Reply With Quote
Old 17th Feb 2019, 8:14 pm   #32
Bazz4CQJ
Nonode
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 2,193
Default Re: Looking to build RF detector probe for low frequency

Quote:
Originally Posted by Argus25 View Post
If a Tunnel diode is used in reverse (called a back diode) it acts like a diode with zero forward voltage drop, but it only works for low level signals because its effective PIV rating is quite low.
Ever since the 2016 thread on diode detectors, it's been a subject which I keep a constant eye on, and some time ago, I found references to the use of tunnel diodes (TD) as detectors for low signal levels. Unfortunately, it seems that most of the papers I found (typically from the 1960's) looked the question from a theoretical standpoint. None of the papers I found and filed have any practical descriptions of such detectors.

One reason for my interest in TD's is simply that I have some, salvaged from a computer board in the early 1970's, but for those who do not have them in their junk boxes, quite a range of ex-Soviet TD's are available cheaply on-line.

If it's the case that the TD's have limited signal handling capacity, can't this be made a virtue by way of needing to reduce the coupling of the TD to the circuit under test, and so imposing minimum loading on that?

B
__________________
We know there are known unknowns. But there are unknown unknowns – ones we don't know we don't know.
Bazz4CQJ is offline   Reply With Quote
Old 18th Feb 2019, 6:32 pm   #33
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,545
Default Re: Looking to build RF detector probe for low frequency

I'm not sure I'd recommend the Haig circuit if demodulation is also required because I think there will be a fair bit of distortion in the demodulated signal. At small signal levels the distortion will be in the diode detector and at larger signal levels (i.e. a large waveform at the drain in high gain mode) the JFET will become non linear.

I guess it would still be OK if you are content to accept something like a wonky sinewave from the detector. However, if reasonable fidelity is required then I'd try a different approach. Either that or try and improve the Haig circuit if you want to reuse as many parts as possible.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Old 19th Feb 2019, 5:38 pm   #34
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,545
Default Re: Looking to build RF detector probe for low frequency

I know IsquaredR only wants to use this probe up to 3 or 4MHz but if anyone else wants to use the Haigh probe as described in the Haigh article then I'd advise caution if you want to avoid disappointment. The second paragraph in the Haigh article implies that this probe was designed for use (in high gain mode) in comms receivers with a high first IF. The article implies it could be used to probe the response of the first IF filter. If this first IF is typically up at 45MHz and has an impedance of a few hundred ohms then I'd argue that the Haigh probe isn't fit for purpose in this respect because it will load the circuit under test quite a bit. It would be just about OK to use in a 50R circuit at 45MHz but I wouldn't recommend it.

I get the impression that Haigh simply assumes or hopes that the input impedance of the JFET will be high at VHF when in high gain mode. But any kind of basic analysis (or measurement) will prove otherwise. It isn't as simple as assuming that JFETs are high impedance devices so the input loading can be modelled by 1Meg in parallel with a few pF.

The J310 JFET has 2pF of D-G capacitance, a Gm of maybe -0.01 and the drain load is 1000 ohm. So there will be Miller Effect at play here. The Miller capacitance could typically be 22pF. With a Gm of -0.01 and drain load of 1000 ohm the Rs of this capacitance could be around 91 ohms based on this ultra crude model.

The parallel equivalent of (91R in series with 22pF) at 45MHz is about 375R in parallel with about 17pF and this is roughly what will be seen at the input of the JFET in the Haigh circuit at 45MHz in high gain mode. This is a very crude and nasty model but it will get pretty close to reality at 45MHz I think. However, in reality, there will be additional parasitic resistances and capacitances within the JFET so the impedance will be slightly lower than this but this basic analysis will be close enough I think. So I would argue that the probe will detune and load filter circuits quite badly at 45MHz unless the probe is used at very low impedance points in the circuit.

This Haigh circuit seems to be the latest in a series of JFET buffered diode detectors that have appeared on the forum in recent years. All the ones I've seen have been flawed in various ways and my advice would be to be very wary of what is promised in magazine based projects like this.
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 19th Feb 2019 at 5:54 pm.
G0HZU_JMR is offline   Reply With Quote
Old 19th Feb 2019, 6:03 pm   #35
stuarth
Hexode
 
Join Date: Feb 2010
Location: Heysham, Lancashire, UK.
Posts: 311
Default Re: Looking to build RF detector probe for low frequency

I think the hand capacity effect could be due to the long ground lead.

There is a loop, a one turn coil, formed by the circuit from the signal source to the probe input, then through the probe and its ground lead and the connection back to the signal source. This loop will have resonances which will cause peaks and troughs in the frequency response. The size and area enclosed by the loop should be minimised - see GMB’s post 21.

For high impedance scope probes, Tektronix (and others) supply a range of clips and leads. On fast digital signals, you can see the effect of a long earth lead as aberrations, overshoot, undershoot, ringing etc, on the fast edges, and if it is affects what you’re trying to measure, you can see the problem and take steps to fix it, ie to get a clean edge.

My default is a 3” lead, with a little probe tip ground pin used when looking for problems on fast edges. A long 12” ground lead is only really for audio.

For RF measurements, you will just have a wrong amplitude, or an amplitude which varies as you move the ground lead, but it’s hard to see what is “right”, you don’t have the equivalent of the “clean edge”.

The fact they you are seeing hand capacity effects shows that there is a problem. It’s worth trying the shortest ground wire possible, connected to ground as close to the signal source as possible.

Stuart
stuarth is offline   Reply With Quote
Old 21st Feb 2019, 5:14 pm   #36
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,545
Default Re: Looking to build RF detector probe for low frequency

There's also a long internal connection from the probe tip to the JFET PCB and it runs parallel to the battery and other wires. I have no idea how much this actually matters (in terms of stray pickup) when probing LF circuits but it might be worth trying a few experiments inside the box. Maybe swap things around internally so the probe tip connection is much shorter.

I'm still not impressed by the Haigh circuit design for the reasons already listed, but if it proves to be a useful tool at lower frequencies then I guess this doesn't matter.

One neat way to use this tool (with a Tek 465 scope) would be to connect the CH1 output on the back of the scope to an audio amplifier and speaker. I've not tried it but I'd expect it to produce audio from the speaker as the IF signal path is probed. The 465 CH1 output has a low impedance and it has a hard limit so it's unlikely to produce objectionable (hurtful?) levels of audio when probing higher gain stages by accident. But that's just a guess. I've used the CH1 output to drive a 50R spectrum analyser many times and the CH1 output can't damage the analyser no matter what is input to the CH1 Y input of the scope and no matter what gain setting the scope has.
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 21st Feb 2019 at 5:42 pm.
G0HZU_JMR is offline   Reply With Quote
Old Yesterday, 9:06 pm   #37
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,545
Default Re: Looking to build RF detector probe for low frequency

Quote:
Originally Posted by ianevans View Post
I've just seen this thread after now joining the forum. I have a number of probes from Eico, and one Heathkit I picked up recently, boxed with instructions and all the parts as an unmade kit. I'm thinking I've got the ones built and all the circuits, and of course this Heathkit. Is anyone interested in this being put up on this thread?
Cheers and hello fellas, Ian Evans
Hi Ian, I just reread the thread and I think we all missed that this was your first post on the forum!

So first of all welcome to the forum

I'd recommend that you start a fresh thread for your probe kits. I'd be interested to see the circuits and so would a few others as there is always interest in any kind of active or passive RF probe.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Reply

Thread Tools



All times are GMT +1. The time now is 12:29 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 - 2019, vBulletin Solutions, Inc.
Copyright ©2002 - 2018, Paul Stenning.