Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently Jul Mon 13, 2020 3:43 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 17 posts ] 
Author Message
 Post subject: RF Probes
PostPosted: Jun Mon 29, 2020 4:50 pm 
Member

Joined: May Wed 20, 2020 5:06 pm
Posts: 81
I was doing some experimenting and wanted to get some feedback from guys more experienced than me. I did this on a simple but working 4 band radio. I wanted to get a feel for what to expect when a radio does have high band problems. Again this radio is working ok.

This radio’s band 4 is its high end at 10.5 to 30 MHz. It doesn’t have an RF amp. It has a 12BE6 converter circuit on the front end. The schematic says pin 1 on the 12BE6 should have -7V. The notes say take readings in band 1.

I tried monitoring pin1 with both a Tek 465 scope and a Heathkit IM-28 VTVM. I was using a standard (non-RF) probe on the VTVM. By that I mean the probe has a 1 megohm resistor in series on DC and straight through on AC. I was using a 10x probe on the scope.

First approximate scope readings peak to peak, band 1 = 25V, band 2 = 20V, band 3 = 15V, and band 4 flat line. So, the scope loads the OSC down too much on the high band.

Next, I connected the scope probe through a separate 1 megohm resistor. Now the scope doesn’t kill the OSC and displays it. Then I tried a 10K resistor. Both resistors show an undistorted peak-peak waveform of about 800mv.

Second the VTVM, band 1 = -9.8V, band 2 = -9.0V, band 3 = -5.2V, and band 4 = -2.0V. So, the VTVM with a standard (non-RF) probe doesn’t kill the OSC.

My question is where does the need for an RF VTVM probe come into play, at much higher frequencies like above 100 MHz? Or is it just that this radio has a strong enough OSC that I can use a non-RF probe? Does a person really need RF scope or VTVM probes working on radios below 30 MHz?


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jun Mon 29, 2020 4:54 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3534
Location: Seattle WA US
The RF VTVM probe is needed when you want to measure AC voltages at a frequency above the frequency range of your VTVM without the added probe. This frequency limit varies from VTVM to VTVM design.
-Chuck


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jun Mon 29, 2020 5:33 pm 
Member

Joined: Jun Sun 15, 2014 5:37 pm
Posts: 471
Location: Montreal, Quebec
The HP-410B can measure AC voltage up to UHF. But most VTVMs are good through not.much more than audio frequencies.

A key consideration is that the HP-410B not only has a fancy tube to detect, but it puts it in the probe. You don't want the usual meter probe carrying RF back to the meter where it's detected. That will load down the circuit under test.

RF probes are generally simple, a coupling capacitor, a diode, and a load resistor. The AC detector inside a vtvm will be about the same. Making them is easy, though then you need to calibrate the system if you want more than a relative reading.


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jun Mon 29, 2020 5:37 pm 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 4745
Location: Littleton, MA
Gilligan wrote:
First approximate scope readings peak to peak, band 1 = 25V, band 2 = 20V, band 3 = 15V, and band 4 flat line. So, the scope loads the OSC down too much on the high band.

Next, I connected the scope probe through a separate 1 megohm resistor. Now the scope doesn’t kill the OSC and displays it. Then I tried a 10K resistor. Both resistors show an undistorted peak-peak waveform of about 800mv.

Second the VTVM, band 1 = -9.8V, band 2 = -9.0V, band 3 = -5.2V, and band 4 = -2.0V. So, the VTVM with a standard (non-RF) probe doesn’t kill the OSC.

My question is where does the need for an RF VTVM probe come into play, at much higher frequencies like above 100 MHz? Or is it just that this radio has a strong enough OSC that I can use a non-RF probe? Does a person really need RF scope or VTVM probes working on radios below 30 MHz?

You are measuring two different things.

On your scope, you are measuring the peak-to-peak voltage of the oscillator signal. (Presumably the scope input was set to AC coupling.)

On your VTVM, you are measuring the grid-leak DC bias on the oscillator.

Even with an RF probe for your VTVM, you wouldn't be able to measure the AC signal on the oscillator grid. VTVM RF probes have a relatively low input impedance, and so would kill the oscillator.

_________________
Steve Byan https://www.byan-roper.org/steve/steve-at-play/


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jun Mon 29, 2020 6:43 pm 
Member

Joined: May Wed 20, 2020 5:06 pm
Posts: 81
I guess my typing got ahead of my poor brain :lol: Not sure where I was going with that last question about using an RF probe on a DC voltage.....

BTW, I bought that VTVM from you (I think) at last year's Brookline meet. You did a nice job rebuilding it. The probe you built still works great, too!


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jun Mon 29, 2020 7:51 pm 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 4745
Location: Littleton, MA
Gilligan wrote:
BTW, I bought that VTVM from you (I think) at last year's Brookline meet. You did a nice job rebuilding it. The probe you built still works great, too!

The Heathkit IM-28 with my homebrew switchable probe? Yeah, that was me. I'm glad you like it.

I documented its restoration on my blog:
https://www.byan-roper.org/steve/steve- ... stora.html

and here on ARF:
https://antiqueradios.com/forums/viewtopic.php?t=365180

_________________
Steve Byan https://www.byan-roper.org/steve/steve-at-play/


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jun Mon 29, 2020 8:48 pm 
Member

Joined: Jun Fri 19, 2009 6:34 pm
Posts: 9587
Location: Long Island NY
The difference is, the lowest scale on a typical bench VTVM is 1.5 volts. Most RF voltmeters can measure down to the millivolt or microvolt level. Most are fairly broadband; depending on make and model, they may be usable from say, 100-kHz up to 1-GHz.

Most service bench VTVMs with diode(s) in the case of the meter are good for about 3 MHz to 4 MHz with considerable loading at higher frequencies due to the probe cable. In some VTVM models the diodes are in the probe; those will typically remain accurate until you hit 200 MHz to 300 MHz. The aforementioned HP 410 series of meters had custom probes with microwave diodes that remained accurate to about 700 MHz and still provided usable indications to perhaps 1.2 GHz.

As for oscilloscopes, the probe and input capacitances become a big load as the frequency is increased. One megohm in parallel with 24 pF is one megohm at DC, but at 30 MHz it looks like 221 ohms. Yes a X10 probe will isolate the scope input from the circuit you are trying to test to a certain extent, but that is offset by the capacitance of the probe itself.

When you use a VTVM on DC voltage with a 1-megohm tip resistor, it becomes possible to measure the DC voltages in oscillator circuits without shutting them down. The tip resistor isolates the circuit under test from the effects of the capacitance in the probe cable and meter. What you are seeing is fairly typical of multiband radios. Pin 1 of a 12BE6 is the oscillator grid, so it stands to reason that the more vigorously a tube is oscillating, the greater this voltage will be. Since oscillators are essentially non-linear circuits, rectification takes place and a DC meter will work. Thing is, as the frequency is increased it gets harder for the tube to oscillate, so it is normal for the voltage to fall off as you go up the ranges,

Bets regards,
Chris

_________________
"Hell, there are no rules here--we're trying to accomplish something!"

Thomas A. Edison


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jun Mon 29, 2020 10:10 pm 
Member

Joined: May Wed 20, 2020 5:06 pm
Posts: 81
Thanks Chris. That was very helpful. I was thinking about buying or building an RF probe, but I guess I really need one for the kinds of things I do. I bought a radio that has a weak 15-30 MHz band. While I'm waiting on it to arrive, I thought I would try to get a better understanding of how to troubleshoot OSC sections with my trusty old Tek 465.


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jun Tue 30, 2020 12:02 am 
Member

Joined: Jun Fri 19, 2009 6:34 pm
Posts: 9587
Location: Long Island NY
Not sure I understand what you mean by "the kinds of things." Are you looking to repair or restore old radios, modify them or improve their performance, or delve into the technical aspects of how they work? If you simply want to find out if an oscillator is running, a Tek 465 can show you whether or not you've got some kind of signal when you hold the probe near the oscillator coil. You can do the same thing with a small digital radio for that matter--and the radio will tell you what frequency you are on. The DC VTVM with the one-meg probe tip resistor is the classic way of doing it; if the oscillator isn't running you won't get the expected voltage on the grid, nor will the plate voltage be where it should be. You can pretty much tell if the oscillator is running at the right level by doing a sensitivity test on the radio; if the oscillator is weak there won't be as much conversion gain as there should be and the radio won't get distant stations that others can pick up.

Of course if you want to dig deeper into radio design and/or put numbers to your test results you'll need to develop the methodology, which may require additional test equipment depending on how far you want to go.

_________________
"Hell, there are no rules here--we're trying to accomplish something!"

Thomas A. Edison


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jun Tue 30, 2020 4:53 am 
Member

Joined: Jan Thu 16, 2020 12:29 am
Posts: 583
It is actually quite difficult to make a good RF probe that can measure low voltages.

Even schottky diodes require about 100mV to get started and in the early phases have square law rectification, though there are precision rectifier circuits with op amps where you can linearize them and overcome the forward drop.

However all "active" probes have a limited dynamic range compared to just diodes alone.

HP made a very good passive diode RF probe with two diodes, one of them compensated for the non linearity of the other, it was clever.

The bottom line is if any of the RF is amplified before it is detected, there needs to be a satisfactory attenuator at the input, and when you have designed it, to be a wide band system, you will have egg on your face because you will have merely re-created the input system on an oscilloscope and realized you have re-invented a wheel and should have bought a good scope in the first place. I use a Tek 2465B (400MHz bandwidth)

There are other ways to detect RF at very low levels, for very small level signals, a Tunnel diode used in reverse (called a Back diode in that application) works.

But probably the most brilliant method I have seen was featured in Horowitz & Hill's book, the Art of Electronics. They arranged to drive the rectifier diodes with a current source, rather than a voltage source. The output voltage merely rises to whatever value is required to create a diode current proportional to the input voltage. On account of this, the diode's forward voltage drop and square law non-linearity vanish- genius.


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jul Sat 04, 2020 3:43 pm 
Member

Joined: Jan Thu 16, 2020 12:29 am
Posts: 583
Well it is surprising that this thread went quiet, since RF probes are a very interesting topic. I have attached the HP design I was talking about that didn't generate any interest. It does have a very wide dynamic range and an amazing bandwidth. Whenever I mention a clever circuit, nobody ever asks to to see it. I'll wait for someone else to copy the page from H & H's textbook on the current source driver to linearize detector diodes.

And I didn't even get to talk about RF detector devices like the the UTD-1000 or the inexpensive RF measuring IC's coming out of China.


Attachments:
RFprobe.jpg
RFprobe.jpg [ 364.39 KiB | Viewed 205 times ]
Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jul Sat 04, 2020 3:58 pm 
Member

Joined: May Thu 14, 2015 4:15 pm
Posts: 2252
Location: Dallas, TX
ACORNVALVE wrote:
Well it is surprising that this thread went quiet, since RF probes are a very interesting topic. I have attached the HP design I was talking about that didn't generate any interest. It does have a very wide dynamic range and an amazing bandwidth. Whenever I mention a clever circuit, nobody ever asks to to see it. I'll wait for someone else to copy the page from H & H's textbook on the current source driver to linearize detector diodes.

And I didn't even get to talk about RF detector devices like the the UTD-1000 or the inexpensive RF measuring IC's coming out of China.

That's interesting. Please add more info on other designs. I'm always up to learn more.

_________________
Tim
It's not the Destination, It's the Journey.


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jul Sat 04, 2020 4:08 pm 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 4745
Location: Littleton, MA
The Fluke 85RF probe has a similar design to the HP 11096B:

Attachment:
Fluke 85RF schematic.png
Fluke 85RF schematic.png [ 44.23 KiB | Viewed 200 times ]


Attachment:
Fluke 85RF parts list.png
Fluke 85RF parts list.png [ 101.23 KiB | Viewed 196 times ]


The Fluke 85RF II is somewhat different:

Attachment:
Fluke 85 RF II schematic.png
Fluke 85 RF II schematic.png [ 45.21 KiB | Viewed 200 times ]


There are a couple of interesting designs in Hayward, Campbell, and Larkin's Experimental Methods in RF Design. I experimented with the earlier version of Hayward's biased balanced diode design that he published in Solid State Design for the Radio Amateur. I breadboarded it and noticed that the offset is very thermally sensitive. The two diodes have to be tightly thermally coupled. In retrospect this shouldn't have been surprising, but it isn't noted in either description of the circuit.

Then there are the balanced-diode-in-the-probe designs as in the Ballantine RF millivoltmeters and the HP 411A.

All of these RF voltmeter probes suffer from a fairly low input impedance, especially as the frequency increases. Hence the usual solution is to make substitution measurements using a matched-impedance RF power detector. The Analog Devices AD8307 is an excellent basis for a home-brew RF power meter.

_________________
Steve Byan https://www.byan-roper.org/steve/steve-at-play/


Last edited by stevebyan on Jul Sat 04, 2020 4:18 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jul Sat 04, 2020 4:16 pm 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 4745
Location: Littleton, MA
Notimetolooz wrote:
That's interesting. Please add more info on other designs. I'm always up to learn more.

Here are two designs from Horowitz and Hill's Art of Electronics, 2nd Edition. (BTW, the third edition omits much of the interesting RF stuff in the 2nd edition.)

Attachment:
Wideband linear detector.png
Wideband linear detector.png [ 311.78 KiB | Viewed 196 times ]


Attachment:
self-compensating diode detector.png
self-compensating diode detector.png [ 231.89 KiB | Viewed 196 times ]

This circuit is from one of the HP counters, I don't remember which one.

_________________
Steve Byan https://www.byan-roper.org/steve/steve-at-play/


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jul Sat 04, 2020 10:50 pm 
Member

Joined: Jan Thu 16, 2020 12:29 am
Posts: 583
stevebyan wrote:
All of these RF voltmeter probes suffer from a fairly low input impedance, especially as the frequency increases. Hence the usual solution is to make substitution measurements using a matched-impedance RF power detector. The Analog Devices AD8307 is an excellent basis for a home-brew RF power meter.


Silicon Chip recently published (July 2020 page 27) a review of a 100kHz to 500MHz RF power meter from Banggood in China, for less that $50. It uses the AD8037 and a meter module with an LCD display. Its amazing how they got the cost so low.

One of my favorite devices which I use to monitor RF levels in Television Standards converters (for 625-405 line conversion with the original UK RF carriers) is the UTD-1000. These work beautifully well , but they are costly. Edit: I have attached a photo of a unit I built, it used an OP amp on the output of the UTD-1000, in this case to drive an analog meter.

https://www.datasheets360.com/pdf/-1762504055471767690

I think the Horowitz & Hill current drive circuit would make an excellent RF probe and a fet could be easily added to the input to lift the input Z.


Attachments:
UTD1000.jpg
UTD1000.jpg [ 277.29 KiB | Viewed 160 times ]
Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jul Sat 04, 2020 11:26 pm 
Member
User avatar

Joined: May Wed 18, 2011 2:40 am
Posts: 4745
Location: Littleton, MA
ACORNVALVE wrote:
HP made a very good passive diode RF probe with two diodes, one of them compensated for the non linearity of the other, it was clever.

Like The Fluke 85RF, the HP 11096B uses a germanium diode as the detector and a silicon diode for the linearity compensation.

Attachment:
HP 11096B parts list.png
HP 11096B parts list.png [ 375.79 KiB | Viewed 154 times ]

_________________
Steve Byan https://www.byan-roper.org/steve/steve-at-play/


Top
 Profile  
 
 Post subject: Re: RF Probes
PostPosted: Jul Sun 05, 2020 12:39 am 
Member

Joined: Jan Thu 16, 2020 12:29 am
Posts: 583
Notice how in the H & H current source circuit, the collector load for the input transistor is a current source. These transistors can actually be replaced with an inductor that acts like a current source (because it opposes a change in current) and has a very high Z at the operating frequency. If the circuit is reconfigured like that it confuses most who then think it is merely an RF amplifier, with a choke as a load driving the rectifiers, and is no different an any other situation where you might see this, say at the detector output in a radio. But it is not.

If you look at H & H's circuit the diode load is 100R. But the current source (or inductor if one was substituted) has a very high Z. So it is a circuit with an intentional gross impedance mismatch, which you do not see in conventional detector design. So in some cases, gross impedance mismatching is beneficial and as it turns out, at an RF detector, this is one of those rare occasions, but not many manufacturers, of radios for example, recognized or exploited this interesting concept to improve a detector's low signal level performance.

I once built a "Supadetector" based on this idea from H&H, and used a large inductor for the current source, to make a module to drop into a receiver to improve its performance for low level signals. It is interesting, that for this circuit, the low signal level performance and linearity improves with increasing loads or mismatch within the limits of overall reduced signal output. The un-bypassed emitter resistor helps ensure that the transistor's collector current is linearly related to the input voltage. As can be seen this makes a useful detector down to an RF input of 10mV peak, much better than a 1N60.


Attachments:
SUPADETECTOR.jpg
SUPADETECTOR.jpg [ 729.01 KiB | Viewed 142 times ]
Top
 Profile  
 
Post New Topic Post Reply  [ 17 posts ] 

All times are UTC [ DST ]


Who is online

Users browsing this forum: Tony F and 15 guests



Search for:
Jump to:  


































-->


Privacy Policy :: Powered by phpBB