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 4th Nov 2018, 2:04 am   #41
Radio1950
Pentode
 
Join Date: Apr 2017
Location: Brisbane, Queensland, Australia.
Posts: 140
Default Re: RF power meter calibration

Bazz4CQJ,

The coax is curved for physical relief of thermal effects; the little tags are fragile.
Other methods may be better.
Caution these devices have BeO.
The Anaren RFP250N50 datasheet is good for practical mounting advice.
You can also sometimes buy these devices "used" at very low prices on the usual budget website.
I have another low power (5W) load using a Florida RF 50W device; it was bought as "used".
You might find that "used" high power 100 ohm devices (ex Wilkinsons) are cheap; just use in parallel.

I wonder about using a thermistor on such a large heatsink, as there will be considerable time delay with the temp rise for measurement, no matter where it is physically placed.
Radio1950 is offline   Reply With Quote
Old 4th Nov 2018, 2:27 am   #42
Radio1950
Pentode
 
Join Date: Apr 2017
Location: Brisbane, Queensland, Australia.
Posts: 140
Default Re: RF power meter calibration

Just going back to the original request.

Steve,
I cant help thinking that the info in posts #2 and #3 is most helpful for home cal at HF.


We still dont know what type of meter you have for cal??


If you want a very good paper on some of this, HP published this -

"Fundamentals of RF and Microwave Power Measurements
Application Note 64-1B".

And a few photos of my power meter and some 1960s Vintage gear-
1 - I calibrated this power meter with Sig Gen, CRO, HF TX.
2 - Inside a Vintage Tenco SWR meter
3 - Inside a Vintage Osker SWR200 meter.

Both the Tenco and Osker meters test at 20% power accuracy from 5-30 MHz, and are hopeless below 4 MHz as the line coupling is too low (too short a line).
Attached Thumbnails
Click image for larger version

Name:	1 - HF Power Meter  DSC03545.jpg
Views:	17
Size:	47.6 KB
ID:	172097   Click image for larger version

Name:	2 - Tenco SWR DSC03556.jpg
Views:	18
Size:	42.8 KB
ID:	172098   Click image for larger version

Name:	3 - Osker Block 200 SWR DSC03558cr.jpg
Views:	20
Size:	67.9 KB
ID:	172099  
Radio1950 is offline   Reply With Quote
Old 4th Nov 2018, 3:49 am   #43
Bazz4CQJ
Nonode
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 2,066
Default Re: RF power meter calibration

Quote:
Originally Posted by Radio1950 View Post
Caution these devices have BeO. The Anaren RFP250N50 datasheet is good for practical mounting advice.
When I took a look on-line, the only devices I could find were from Chinese suppliers (I think I searched for 'flange resistor', maybe another term would have been better). There was a 150W made by Anaren, but the datasheet shows it to be aluminium nitride rather than BeO, and also a 250 device, but I couldn't spot the make of that from the picture, but that too is aluminium nitride. Both types were very inexpensive (~£1). I just ordered a few 250W pieces, though wish I played safe and also got some of the Anaren.

I'm planning to put either the thermocouple or the thermistor directly under one of the screws holding down the resistor. Alternatively, I do have a good quality infra red instrument, so I'll be able to experiment. The junk box has offered up a sheet of copper and a heat sink quite similar to the ones you've used.

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 4th Nov 2018, 4:06 am   #44
Terry_VK5TM
Heptode
 
Join Date: Oct 2010
Location: Tintinara, South Australia, Australia
Posts: 588
Default Re: RF power meter calibration

There was at one stage the proper termination pcb's to go along with the RFP resistors for sale on the usual sites.

Similar to the drawing attached, they were single sided and cost more than the resistor itself.

Piece of scrap pcb and a sharp knife would knock one up in no time and for almost nothing.

(For reference, the track is 2.8mm wide as is the spacing between track and ground area and the coax needs to be terminated both sides of the track. Spacing based on average Er of FR4 material).
Attached Thumbnails
Click image for larger version

Name:	Termination pcb.JPG
Views:	16
Size:	31.4 KB
ID:	172106  
__________________
Terry VK5TM
https://www.vk5tm.com/
Terry_VK5TM is offline   Reply With Quote
Old 4th Nov 2018, 1:25 pm   #45
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,493
Default Re: RF power meter calibration

It's definitely worth measuring the load to see how accurate it is, especially up at 29MHz when using the diode or scope method.

With the scope or diode method the following will apply:

A 10% error in the load (eg 55 ohm) will mean about a 10% error in the power calculation.
A 3% voltage error with the old analogue scope would mean about a 6% error in the power calculation.
A -30dBc harmonic in the source could mean about +/-6% uncertainty in the power calculation.
A -40dBc harmonic in the source could mean about +/-2% uncertainty in the power calculation.

Using a x10 scope probe is a bit dodgy because it will have to be compensated correctly and the response known to be flat to 30MHz and beyond. If it peaks up sharply at VHF then it could make the error conributions worse from harmonics up at VHF. Also, I wouldn't want to use a x10 scope probe at 100W at 29MHz because a typical low cost x10 scope probe will be starting to look like a dissipative load by 29MHz.

Don't use a scope directly with a tee piece above about 10-15Vrms if operating across the HF band. I would advise to never connect an HF transmitter directly to a scope even if it is only a few watts. It's not worth the risk, especially if the load accidentally disconnects and the voltage at the scope input goes up. The scope will look like a dissipative load by 29MHz and parts inside could get stressed. It gets much worse by 50MHz.
If you use the diode detector method then use a decent Schottky diode like the HP2800 or 1N5711. Don't use one of the cheap/classic germanium glass diodes, the detector performance up in the linear region will usually be inferior.

However, even if you ignore all the advice above you might get away without damaging your scope and it probably doesn't matter (in the real world) if the error contributions above meant you actually adjusted the meter to make it slightly worse than it was originally Maybe it would be a good idea to measure the trimmer resistors in the meter with a DMM before adjusting them. So at least it could be put back where it was if things don't work out.
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 4th Nov 2018 at 1:42 pm.
G0HZU_JMR is offline   Reply With Quote
Old 4th Nov 2018, 7:10 pm   #46
Bazz4CQJ
Nonode
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 2,066
Default Re: RF power meter calibration

Quote:
Originally Posted by G0HZU_JMR View Post
With the scope or diode method the following will apply: Using a x10 scope probe is a bit dodgy because it will have to be compensated correctly and the response known to be flat to 30MHz and beyond. If it peaks up sharply at VHF then it could make the error conributions worse from harmonics up at VHF. Also, I wouldn't want to use a x10 scope probe at 100W at 29MHz because a typical low cost x10 scope probe will be starting to look like a dissipative load by 29MHz.
The scope will look like a dissipative load by 29MHz and parts inside could get stressed. It gets much worse by 50MHz.

If you use the diode detector method then use a decent Schottky diode like the HP2800 or 1N5711. Don't use one of the cheap/classic germanium glass diodes, the detector performance up in the linear region will usually be inferior.
Could we break this down a little, and as per OP, stick to 100W at no more than 30MHz, at least to begin with.

Re the Power Meter (PM) load; the 50R flange resistors, which are intended for terminator applications, seem to be the consensus choice for a new "amateur build" PM, and for sure, the actual value of that component needs to be checked as accurately as possible (probably, in both the "as received condition" as well as after some period of ageing at elevated temperature (100'C max spec).

Coming directly off the 50R load (no T pieces in the feed), we have an attenuator feeding either a diode probe or a scope. There seems to be consensus that the capacitance involved in using a scope complicates the issue; my understanding here is that the capacitance is effectively reducing the resistive value of the end of the attenuator, so giving a false, low reading which is also dependent on frequency?

If that is correct, is the optimum design for the attenuator in this role, the one that gives just enough resistive attenuation so that the scope can handle the voltage safely, so the effects of C are minimised?
What (and why) would the optimum sensitivity setting of the scope be for such a measurement?

Looking at the diode detector option, I had anticipated that the low capacitance of the old point contact diodes would make them attractive for this role, and we are not short of power, so we can easily put enough voltage on them to be sure we are well in the linear region.

Re harmonics, impossible to assess in the absence of a spectrum analyser?

I'm sorry if I'm being slow to appreciate the subtleties here, but judging by the number of articles on the internet about RF Power Measurement, which have been criticised here, and the old previous thread on this forum, it does seem to be a subject where many have underestimated the problem of RF power measurement. But, it's an interesting subject and of significant practical use for anyone who is licenced!

As an aside, quite lot of amateurs are using ATU's made by MFJ. These are certainly built to a tight budget, but generally seem to be assessed as OK for 100W. Most models have a Power Out/SWR meter fitted. I searched the internet yesterday looking for comments on how accurate those meters might be; found absolutely nothing!

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 4th Nov 2018, 8:00 pm   #47
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 6,502
Default Re: RF power meter calibration

Quote:
Originally Posted by Bazz4CQJ View Post
As an aside, quite lot of amateurs are using ATU's made by MFJ. These are certainly built to a tight budget, but generally seem to be assessed as OK for 100W. Most models have a Power Out/SWR meter fitted. I searched the internet yesterday looking for comments on how accurate those meters might be; found absolutely nothing!

B
You need to define the mode used in order to get any sense of power.

"Key-down" on a CW transmitter might give you 100 Watts for a while; the same transmitter delivering un-speech-processed SSB voice might only deliver 15 Watts or so averaged over a few seconds.
This is why ham transmitters often overstate their output on SSB; the duty-cycle is low so they can get away with heatsinks that allow them to deliver '100W PEP' in sideband service but only 25 Watts in continuous-carrier modes like FM, double-sideband-and-full-carrier AM, or RTTY/Digital modes.
G6Tanuki is offline   Reply With Quote
Old 4th Nov 2018, 8:26 pm   #48
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,493
Default Re: RF power meter calibration

Quote:
Looking at the diode detector option, I had anticipated that the low capacitance of the old point contact diodes would make them attractive for this role, and we are not short of power, so we can easily put enough voltage on them to be sure we are well in the linear region.
I recommend using a 1N5711 Schottky diode because they have consistent performance in terms of efficiency and leakage up into VHF. A randomly selected (ebay?) Ge diode is likely to cost more and perform worse than the 1N5711 here so why take the risk

Quote:
There seems to be consensus that the capacitance involved in using a scope complicates the issue
I'm not sure a scope is the right tool up at 29MHz if the aim is to check the accuracy of a power meter. Also, I'd be concerned about damage to the scope.

There will obviously be some variation (in input impedance) between scope models and x10 scope probes but I'd expect both the scope and the probe to look resistive (Rs) as well as capacitive (Cs) across the HF band and the problem with using a raw scope input up at 29MHz is that the effective shunt resistance (Rp) of the scope begins to drop quite alarmingly by 30MHz. The load VSWR is also degraded by the capacitance.

By design, many scopes have some deliberate ESR close to the input connector. This is probably there for more than one reason but it is often needed to help offset some negative resistance in the active follower stage. So the scope might have a 40-50 ohm resistor in series at the input BNC. So there could be 40 ohms net ESR at the input and this will cause dissipative issues up by 29MHz (the equivalent Rp drops a lot by 29MHz) and it will degrade the accuracy of the 50R load slightly.

So the scope could get hot inside at 29MHz (especially at 100W!) and any power dissipated inside the scope is power registered on the power meter under test (as FWD power) but it won't be accounted for in the V^2/R calculation by the operator. So there could easily be an additional 3% power error hiding here by 29MHz if using a scope. With a test power of 10W at 29MHz there could easily be 0.3W Pdiss inside the scope. Multiply these numbers by 10 for 100W and it gets scary for the scope!

A x10 scope probe will also look like a resistor in series with a small capacitance so there will be some Pdiss in a x10 scope probe up by 29MHz. But it won't be quite as bad as a raw scope input. But the operator would also have to select a x10 probe that was compensated correctly and one that was proven to have a flat response up to 30MHz or so. Otherwise the errors here could be much worse than the subtle Pdiss and VSWR issues caused by using a x10 scope probe.
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 4th Nov 2018 at 8:39 pm.
G0HZU_JMR is offline   Reply With Quote
Old 4th Nov 2018, 9:10 pm   #49
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,493
Default Re: RF power meter calibration

I don't know if this helps but I knocked together a basic excel spreadsheet that predicts the impact of using the raw input of a scope. It has inputs for test frequency, test voltage, the test load of 50R and the Rs and Cs model of a typical analogue scope.

Usually a scope can be modelled fairly well by it's input ESR and its input capacitance. So typical numbers would be 40R for Rs and 21pF for Cs and this model will hold across a wide frequency range like the HF band for example. The spreadsheet calculates the input VSWR, the equivalent Rp of the scope and a few other things.

See below for a test at 30MHz. For a typical scope the internal Pdiss is predicted to be about 0.3W at a 10W Tx power level and the effective load VSWR will be about 1.2:1. So there is a hidden error here of a few percent because of the loading effect of the scope at 29MHz. Rp is predicted to be about 1600 ohms by 30MHz so the scope is effectively putting a small 1600R resistor across the test load at 30MHz and this will create a small error in the power calculation.

This is a simplistic case that assumes a direct connection of the 50R load at the scope input. If a coax tee and patch lead is used then the results will be affected. It's a first stab at this so it may contain nuts but the numbers look about right to me. Note that the series 51R input resistor at the CH1 input of a Tek 465 is only rated at 0.25W and I think this is one part that will get hot inside the scope at 30MHz and a Tx power above about 10W.
Attached Thumbnails
Click image for larger version

Name:	ScopeRp50R.jpg
Views:	10
Size:	64.7 KB
ID:	172174  
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 4th Nov 2018 at 9:18 pm.
G0HZU_JMR is offline   Reply With Quote
Old 4th Nov 2018, 9:34 pm   #50
Bazz4CQJ
Nonode
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 2,066
Default Re: RF power meter calibration

Quote:
Originally Posted by G6Tanuki View Post
You need to define the mode used in order to get any sense of power.
I'm assuming key down CW for maybe 10 seconds.
Quote:
Originally Posted by G0HZU_JMR View Post
I recommend using a 1N5711 Schottky diode because they have consistent performance in terms of efficiency and leakage up into VHF. A randomly selected (ebay?) Ge diode is likely to cost more and perform worse than the 1N5711 here so why take the risk
Quote:
There seems to be consensus that the capacitance involved in using a scope complicates the issue
I'm not sure a scope is the right tool up at 29MHz if the aim is to check the accuracy of a power meter. Also, I'd be concerned about damage to the scope.
I guess the use of the scope stems from the OP (where he?), or earlier posts, but also that most licenced amateurs will have a scope of some kind. Re diodes, I still have quite a few gold-bonded Ge point contacts bought long ago, but I think this is a point of detail .

Ah; now way back in post #26 Wrangler was proposing to use a 30dB attenuator next to the 50R load and I had assumed that the thread had absorbed that suggestion as the means of protecting the scope, but maybe not?

So to sum up, what I think is useful to baseline here is a PM
useful for 100W to 30MHz (as a starter), to be used by driving it with full power CW for up to 10 seconds, and with the load having a closely connected attenuator (~30dB) feeding either a diode probe or a scope.My version would have a temperature sensor(s) too!

Doesn't have to be that concept.

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 4th Nov 2018, 10:01 pm   #51
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,493
Default Re: RF power meter calibration

One niggle is that I don't think a basic diode probe can (easily) be used after a 30dB attenuator because it will not be in the linear region and the impact of the Vdrop of the diode will also affect the calibration of the diode detector. So the calibration of the detector alone could be problematic at small signal levels like this. It could be calibrated at a fixed temperature with a decent sig gen and load but this would require some decent test gear.

The other thing to do is just test it with the scope at lower frequencies and then assume the meter calibration is still OK at 28MHz.

The other scope option is to use a big external attenuator (or coupler) and use a scope that comes with a proper internal (switched) 50R load. If the scope's internal 50R load has decent design integrity the input will look like a resistive 50R with little or no parallel capacitance. So the scope VSWR will be extremely low even at 30MHz and the scope will mimic an almost perfect 50R dummy load. But care would be needed not to damage the internal 50R load with excessive voltage. 5Vrms is a typical maximum.

I think my recently acquired (but 20 year old) HP54825A scope can detect an overload and switch out the internal 50R load very quickly to protect it if such an event happens. But this is just a guess. I spent some time trawling through its internal software with a disassembler this weekend and found this feature in the scope code. But the code serves several scope models and this might not be for my 54825A model and I'm not in a hurry to see if this protection feature really works!
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 4th Nov 2018 at 10:09 pm.
G0HZU_JMR is offline   Reply With Quote
Old 4th Nov 2018, 10:23 pm   #52
Radio Wrangler
Dekatron
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 10,896
Default Re: RF power meter calibration

Yes,

The scope's 50 Ohm termination resistor terminates the cable from the attenuator, so the cable gives a flat frequency response.

The big attenuator reduces the power sent to the scope's termination resistor so it isn't burned out. 5v RMS is about +26dBm or 200mW. Some scopes have 125mW rated terminations, but don't count on them being any more.

Relatively few scopes have protection systems monitoring the input, and even with those known to do so, it is not wise to rely on it. It doesn't always work fast enough to save the resistor or the input amplifiers. If you have protection and know it, still plan things to not need it. Let it be a back-up. Spend a bit of time getting it right and it's still a heluva lot quicker than having to fix a destroyed scope along the way.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline   Reply With Quote
Old 4th Nov 2018, 10:59 pm   #53
Bazz4CQJ
Nonode
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 2,066
Default Re: RF power meter calibration

I suspect that oscilloscopes "vary" a bit. My trusty old Leader scope (kindly given to me by a forum member who was headed for Spain), says it has maximum input voltage of 600V (DC or AC p-p) and an impedance of 1M shunted by 35pF. I wonder if it is the case that it lacks any finesse, but is constructed like a 'brick out-house' to withstand students and poorly-trained field engineers?
OK for me then . I don't recall putting more than a volt or two in to it.

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 5th Nov 2018, 12:18 am   #54
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,493
Default Re: RF power meter calibration

Across various makes/models I think there will definitely be variation in the amount of ESR at the input connector. My Tek 465 is spec'd at about 500Vpkpk (at VLF) and it also states the input impedance is 1Meg in parallel with 20pF.

So based on this it ought to be fairly bulletproof. But in reality there is a series 51R damping resistor at the CH1 and CH2 input. So if you look at the impedance of 51R in series with about | 20pF + 1Meg | in parallel and do this from LF to 100MHz the input impedance changes a lot over frequency.

At RF the 1Meg resistor can be ignored and the input impedance will be dominated by about 40 ohms ESR in series with about 20pF. I think some of the 51R (maybe 10-15R?) gets cancelled by the negative resistance of the active follower and this typically leaves about 40R in series with 20pF.

My spreadsheet predicts the parallel equivalent of this at 30MHz is about 1600R in parallel with 20pF. So the 1Meg Rp shown on the Tek 465 spec sheet is shot to hell by 30MHz

I think my HP 54528A scope is rated at 5Vrms with the 50R mode selected but I don't know what frequency range this applies over. This is about 0.5W. My previous scope was an old 500MHz HP54540C and this had a very accurate internal 50R termination but I don't think it was accurate over the full 500MHz of BW. It was very good up to about 150MHz I recall. I'd expect the 54825A to be very similar as I suspect some of the acquisition board technology in this scope will be similar to the older 54540C. So I might have to derate the 5Vrms spec slightly if I wanted to use the scope up at several hundred MHz with the internal 50R load selected.
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 5th Nov 2018 at 12:30 am.
G0HZU_JMR is offline   Reply With Quote
Old 5th Nov 2018, 12:47 am   #55
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,493
Default Re: RF power meter calibration

Hope it's not going off topic too much but the same ESR issue affects JFETs.
Most people consider a JFET to have many mega ohms input Rp in parallel with maybe 5pF Cp and assume this model holds up into RF.

But a better model for the JFET at RF is maybe 7 ohms Rs in series with 5pF Cs. Even this crude series model will give a fairly decent curve fit with the admittance curves given on the JFET datasheet. OK, you might have to tweak the Rs and the Cs to suit a particular JFET but one fixed value of Rs and Cs will model the small signal input impedance quite well up into V/UHF. The traditional Rp and Cp model will be completely unrealistic by comparison.

So it's worthwhile looking up the input circuit for each scope model to try and predict how much the input ESR will affect the loading of a circuit by 29MHz or so. A lo of scopes are going to look similar to the Tek 465 with Rs of maybe 40R and Cs of 20pF. I think my old Tek TDS2012 is fairly similar and my old Tek 585 was somewhere in this ballpark but I can't remember. I have a couple of classic 100MHz Hameg scopes under the stairs and I can measure these but it won't be for a day or two.

There are various series to parallel impedance converters online so maybe have a play with seeing what happens to 40R in series with 21pF by 30MHz. The equivalent Rp will be about 1600 ohms! At high power this will get hot and also 1600R in parallel with 50R means about 48.5R so that is a hidden 3% error that will affect any power prediction based on V^2/R.

https://www.daycounter.com/Calculato...lculator.phtml
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 5th Nov 2018 at 1:04 am.
G0HZU_JMR is offline   Reply With Quote
Old 5th Nov 2018, 3:33 am   #56
Radio1950
Pentode
 
Join Date: Apr 2017
Location: Brisbane, Queensland, Australia.
Posts: 140
Default Re: RF power meter calibration

There seems to be some unnecessary concerns.

Use of a coaxial “T” as a test port in normal low power HF RF measurements.
Use of a 10x CRO Probe in HF RF measurements.

In practice, in the home workshop, within normal common sense restraints, within manufacturer’s specs, and in the context of the original request by the OP, there is nothing wrong in using either.

I set up a test.

1 - I fed my ICOM 7600 TRX on 3.5 MHz, ATU off, FM, 50W, no mod, into my HF Power Meter on the 50 W FWD scale and terminated with a 50 ohm load.
2 - I read the reflected power on the 2w scale as almost zero.
3 - I place a good BNC “T” at the load input, and repeat, with no visible change to the reflected power.
4 - I place the tip of an "economy", compensated, 10x CRO probe on the “open” end of the coaxial “T”, with probe attached to a 20 MHz CRO, and observe just the very slightest change to reflected power, as expected.
5 - I repeat all at 28 MHz, with just a slight more change to the reflected power, as expected.
6 - I repeat all with a UHF coaxial “T”, with no change to results of previous steps.

Now, I wonder if someone here, with more accurate test equipment, can test the changes as above with a BNC “T” , and report back here in terms of both SWR and Return Loss.

Hands up all those here, with careers in RF, who have never used a BNC “T” as a temporary cable joiner at video or HF?
I have one hand ready.
Radio1950 is offline   Reply With Quote
Old 5th Nov 2018, 3:35 am   #57
Radio1950
Pentode
 
Join Date: Apr 2017
Location: Brisbane, Queensland, Australia.
Posts: 140
Default Re: RF power meter calibration

In practice, I still think that the best method for the home workshop for HF is –

Divide the Power Meter into two parts, the “detector” (or reflectometer) and the “associated DC metering”.

You may have to unsolder the connections.

Sweep the “detector” across your required frequency range with a known good sig gen at say +10 dBM and external DC voltage meter (with suitable DC resistance load), to determine the frequency response of the “detector” at the low, medium and high frequencies. For HF, eg, this could be say 1.8, 7, 28 MHz.

Now that you know how the detector performs across the freq range, you can just use a HF transmitter at various power levels, and a 10x CRO probe, to calibrate your power meter at the low end of frequency, and within the specs of probe and CRO.
You can then rightly assume that it is acceptably accurate at the high end of your freq response of the detector.

All of this is covered by myself and others across many of the posts in this thread.
.
Radio1950 is offline   Reply With Quote
Old 5th Nov 2018, 3:36 am   #58
Radio1950
Pentode
 
Join Date: Apr 2017
Location: Brisbane, Queensland, Australia.
Posts: 140
Default Re: RF power meter calibration

Power Meter Diodes.

Within the context of the original post.
For HF powers over 5 watts, Schottkys eg 1n5711, HP5082-2800 (this is a Vintage Forum!).

For HF powers less than 5 watts, and without amplification, you may get better results with .....
Germanium diodes, OA95 etc, but select pairs for dual reflectometers.

In practice, I have gone full circle, OA95 to 1N5711, BATx, and back to OA95 for better sensitivity at the lower power levels.

Even the VHF/UHF reflectometer section (200-500 MHz) of my Power Meter uses OA95s after trying some BAT42 and BAT85 schottky diodes at low power levels.

In practice, you probably have some adjustment control in the meter for cal, and you can just compensate for the diode sensitivity.

If in doubt, calibrate at the high end of the power scale, where diode sensitivity has less effect.
Radio1950 is offline   Reply With Quote
Old 5th Nov 2018, 3:38 am   #59
Radio1950
Pentode
 
Join Date: Apr 2017
Location: Brisbane, Queensland, Australia.
Posts: 140
Default Re: RF power meter calibration

Having said all the above, I too acknowledge that RF power measurement can be tricky, even at HF.

So what did we use professionally for this task, in a certain govt department that maintained ground Air Traffic Control equipments eg ILS, VOR etc 75-340 MHz and up to 50W ??

This was subject to external audit against ISO9000+ for methodology, accuracy of results, traceable calibration, and staff competence.


Weinschel 10 and 30 dB 50 watt attenuators or similar, and HP432/437 with HP478 sensor etc.
For accurate and practical measurements with portability, you cannot beat this.

(For HF High Power, sometimes we used a rack mounted 75 ohm Bird Termaline (with crossed meter needles!)
Radio1950 is offline   Reply With Quote
Old 5th Nov 2018, 4:32 am   #60
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,493
Default Re: RF power meter calibration

Quote:
Now, I wonder if someone here, with more accurate test equipment, can test the changes as above with a BNC “T” , and report back here in terms of both SWR and Return Loss.
If you just want to measure VSWR then I'm not sure it's necessary to do any extensive tests... A typical x10 scope probe will probably have an Rp of several thousand ohms at 29MHz in parallel with maybe 12pF. I think there could be some variation here though because a x10 scope probe usually has very lossy coax and I'm not sure the Zo of the coax is the same from probe to probe.

But the VSWR should remain quite low with the x10 probe even at 29MHz and it will be dominated by the influence of the 12pF.

Forgive me for posting up yet more simple models but I think you can model a x10 scope probe based on the Zo of its (deliberately) lossy coax. The crudest broadband model for maybe 10-50MHz would be 12pF in series with the Zo of the lossy cable.
So some x10 probes might look like 12pF in series with 50R (across a huge RF bandwidth) others might look like 12pF in series with 100R.

If the probe looks like 12pF in series with 50R then at 29MHz this would mean Rp would be about 4000ohms. So the VSWR of a 50R load in parallel with 4000R and 12pF at 29MHz would be about 1.1:1.

Maybe someone with more experience of playing with x10 scope probes can comment. I don't see them as precision devices and care is required to compensate them correctly and old/tired probes can often have unstable compensation as the probe is handled.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Reply

Thread Tools



All times are GMT +1. The time now is 5:29 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 - 2018, vBulletin Solutions, Inc.
Copyright ©2002 - 2018, Paul Stenning.