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


It is currently Jan Tue 19, 2021 4:12 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 27 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Wed 09, 2020 5:59 pm 
Member

Joined: Jun Sat 18, 2011 5:14 pm
Posts: 252
Location: Washington State 98226
This topic is technically not new... I have been searching the invaluable archives here on the form to learn more about this topic. Here is a recent link to one of the posts regarding the subject.

https://www.antiqueradios.com/forums/vi ... 5&t=356213

I know that a number of folks on the forum will tell me that I don't need a spectrum analyzer to align a AA5 ... I get that and I know I don't need one to align my old Hallicrafters either... but this is a hobby and sometime you want to do things just so you can learn something new.

What I have learned about the subject (so far) is that the input to the newer spectrum analyzers ( the Rigols and Siglents) are all 50 ohm... and that they are very sensitive. So you can not just connect the input of the SA to the IF chain on a tube radio that has a high impedance. If you do you are likely to pull the DUT off resonance and worse; likely destroy the front end of the SA. There are some SA (like the HP 3585A) that have a high impedance input so the impedance mismatch can be solved by using that SA but the front end still needs to be protected.

Mr. Carlsons lab has a homemade device to protect the input of his HP 3585A. That can be viewed here: https://www.youtube.com/watch?v=ETKyKC2Zj-M&t=182s

But his device does not solve the impedance mismatch if your SA has a 50 ohm input.

I ran across a RF active probe sold an Ebay that I think may solve the problem and would like some input from some of the forum gurus that know a heck of a lot more about the subject than I do.

Here is a link to the probe: https://www.ebay.com/itm/RF-Active-Prob ... SwWlNZzA4E

This is the schematic for the probe:
Attachment:
Active Probe Schematic.pdf [38.47 KiB]
Downloaded 54 times


Here are the specs:
Attachment:
Specs.pdf [35.4 KiB]
Downloaded 28 times


Here is the probe itself:
Attachment:
Active probe.pdf [766.16 KiB]
Downloaded 32 times


So the questions are:
1. Will the probe solve the impedance mismatch problem
2. Will the front end of the SA be protected (I would much rather fry the probe than the SA!)

As always thanks in advance for your thought full comments and advice.

John w.


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Wed 09, 2020 6:08 pm 
Member
User avatar

Joined: Aug Tue 28, 2018 9:22 pm
Posts: 2820
Location: Sanford Fla 32771 (USA)
Quote:
but this is a hobby and sometime you want to do things just so you can learn something new.

I have used my scope with my programable sweep generator for IF transformer adjustments, just for giggles.
Always wanted a spectrum analyzer.


Attachments:
1st_IF.JPG
1st_IF.JPG [ 94.88 KiB | Viewed 803 times ]

_________________
Paul of Florida ….I had my patience tested. I’m negative.
https://paulsironhorse.smugmug.com/
Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Wed 09, 2020 6:37 pm 
Member
User avatar

Joined: Dec Sun 21, 2014 6:37 am
Posts: 2351
Location: Portland, TN, USA
I’m certainly not a guru, but looking at the schematic it appears it would solve the impedance issue all right.

I am puzzled by the specs saying Output is DC, as there’s no way DC is sent out - unless those two caps are leaky. :D

_________________
73, de Chuck K4CCW

If talk is cheap, it's because the supply usually exceeds the demand.


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Wed 09, 2020 7:04 pm 
Member

Joined: Jun Mon 24, 2013 3:00 pm
Posts: 1696
Location: Champaign IL 61822
The impedance match is not a problem unless its a wide band sweep. Its
... barely ... a problem with a TV IF alignment at 45 MHz. At a 21 MHz TV IF
its significant. For narrower bandwidths its insignificant.

Couple the probe coax to the test point by stray capacitance ... like 0.5 pF.
Above 5 MHZ start with a step attenuator set for 40 dB in series with you indicator.
This will prevent damage with any device discussed in a vintage forum except
a transmitter or TV horizontal output, for which use 70 dB

At 455 kHZ or lower, you might go to an actual 2 to 4.7 pF cap.
Beware of attaching a 13 pF/ 10 Megohm scope probe directly to RF circuits!
It can ruin alignment and cause oscillators to quit. Clamp it
over a wire's insulation.


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Wed 09, 2020 7:14 pm 
Member

Joined: Jun Sat 18, 2011 5:14 pm
Posts: 252
Location: Washington State 98226
Quote:
I'm certainly not a guru, but looking at the schematic it appears it would solve the impedance issue all right.

I am puzzled by the specs saying Output is DC, as there’s no way DC is sent out - unless those two caps are leaky. :D



My Bad... the specs are for another RF device sold by the same Ebay seller.

https://www.ebay.com/itm/RF-Active-Prob ... SwzilZ2qZH

Those specs are for an RF device with a detector. The descriptions says:

Quote:
This is high input impedance probe intended to measure RF voltage and conversion it to DC voltage. You need just this probe and DMM (Digital Multi Meter) and you can measure small RF voltage. Also you can use probe as RF sniffer, just replace input pin with something like antenna for frequency you want to check. Also you can use this probe with NWT boxes and for many other useful things


So those are not the specs for the RF probe ... sorry for the confusion and good catch

John w.


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Wed 09, 2020 8:00 pm 
Member
User avatar

Joined: Oct Sun 15, 2006 12:57 pm
Posts: 7440
Location: Liberty, Missouri
Hi John, how are you going to protect your probe? Note the specs say 3 volts max in.

Image

Image

========
First and foremost, You want to keep big signals and DC out of your analyzer's input connector either of those will VERY quickly ruin the analyzer's front end.

I don't worry much about impedance mismatches when aligning receivers.

I use a fixed 10X scope probe with my analyzer, attach the probe to the radio's detector input, and usually have plenty of signal to show a decent display during alignment, with virtually no loading of the circuit. A scope 10X scope probe has a 9 Meg Ohm resistor in series with the input lead, so any voltages found in the typical receiver will result in only a few micro-amps of current and poses no danger to your analyzer's input.

Before I had my analyzer and TG, I used a sweep generator and logarithmic detector IC. The displayed response looks identical to what you would see using an analyzer and tracking generator, and its quite a bit cheaper.

Image

Ideally you are looking at the shape of the response of a filter. Sure you want to verify the 3db or 6db corners of the passband, but you also want to verify the stop band frequencies which are typically stated at 60dB down from the pass band. That aside you want the passband to be as flat as possible.

Typically the response of a filter is more or less "fixed" during the design phase, so improving the response by tuning is usually minimal at best, however seeing the effects of tuning is certainly more interesting than just watching the VTVM needle rock back and forth across the peak reading.

_________________
Pics, Bold & underlined text, are usually links.

https://mikeharrison.smugmug.com/

Mike


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Wed 09, 2020 10:24 pm 
Member

Joined: Jun Sat 18, 2011 5:14 pm
Posts: 252
Location: Washington State 98226
Thanks Mike and dtvmcdonald (sorry I don't know your name)... I appreciate your willingness to answer questions. But one question always seems to lead to a couple more.

First... why is that you don't have to worry about the impedance mismatch? Is it because you (we) are mostly interested in the shape (magnitude vs frequency) of the tuned IF circuit and not necessarily an absolute accurate values of the voltages?

Second... Mike says to use a "fixed" 10X probe... does that mean I should not be using my probe that can be switched between 10X and 1X and if so is that because you don't want to accidentally switch it to the 1X position.

Third... dtvmcdonald says that the impedance mismatch is not a problem when we are working in the 455 KHZ range but at higher frequencies the mismatch is a more of a problem.

Quote:
The impedance match is not a problem unless its a wide band sweep. Its
... barely ... a problem with a TV IF alignment at 45 MHz. At a 21 MHz TV IF
its significant. For narrower bandwidths its insignificant.

Couple the probe coax to the test point by stray capacitance ... like 0.5 pF.
Above 5 MHZ start with a step attenuator set for 40 dB in series with you indicator.
This will prevent damage with any device discussed in a vintage forum except
a transmitter or TV horizontal output, for which use 70 dB

At 455 kHZ or lower, you might go to an actual 2 to 4.7 pF cap.
Beware of attaching a 13 pF/ 10 Megohm scope probe directly to RF circuits!
It can ruin alignment and cause oscillators to quit. Clamp it
over a wire's insulation.


Could you elaborate a bit about why / how the frequency effects the impedance mismatch. I do understand that if you couple the input of the SA to the IF circuit with a capacitor then as the value of the capacitor gets smaller (like .5pF) then the impedance that the IF circuit sees (connected to the SA) will be higher. And it makes sense that as the frequency of the IF circuit gets higher the impedance the IF circuit sees (going to the SA) is smaller and could de-tune that circuit and could actually damage the front end of the SA. That's why a 40 dB attenuator is in line. Is my thinking on this correct.

So if I understand you guys correctly -- you wouldn't even bother using the RF probe... and if I use your methodology I shouldn't see any smoke coming out of the SA :(

Thanks for taking the time to educate me and probably several others that are reading along.

John w.


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Wed 09, 2020 11:00 pm 
Member

Joined: Jun Mon 24, 2013 3:00 pm
Posts: 1696
Location: Champaign IL 61822
As to impedance mismatch. We are only measuring shape or RELATIVE
intensity versus tuning, so as you said absolute level does not matter.

But if you use a tiny series cap below a certain frequency response
drops off at 6 dB per octave. For 455 kHZ IFs the bandwidth is so small
that does not matter. What I really meant is that at higher frequencies one
often measure a wide bandwidth. E.G. you sweep the IF of old TV at
21 MHz. Its 6 MHZ wide: 6 dB per octave is quite a slope from say 19 to 25MHz.
The solution, of course, is to go into a high impedance input and use
a capacitive voltage divider. (OR a low impedance input but you would need a big
cap on the bottom side to swamp 50 ohms.)


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Wed 09, 2020 11:06 pm 
Member
User avatar

Joined: Nov Mon 02, 2009 7:01 am
Posts: 4812
Location: Lincoln City, OR 97367
Greetings to John and the Forum:

I usually use a X10 scope probe... with a DC block. ALWAYS use a DC block on the input to a spectrum analyzer when probing in active circuits. I made my own DC blocks with diode clipping on the instrument side to protect against transients.

If you want to observe what happens with an impedance mismatch, connect your scope probe to the tracking generator output. (I assume you are using a spectrum analyzer with a tracking generator; trying to synchronize a spectrum analyzer sweep with that of an external sweep generator is usually an exercise in frustration.)

If you sweep the whole range of the analyzer, you will see a lot of high frequency roll-off and probably some lumps in the response as well. (I am assuming that this is a sweep over probably 1 GHz or so). Now, narrow things down so that you are looking at just a typical IF bandwidth and frequency... say 25 KHz at 455 KHz or 400 KHz at 10.7 MHz. You will observe that when the sweep range is so restricted, it looks pretty flat. Plenty good enough for IF alignment.

I have not had any issues in connecting a scope probe directly to the circuitry. Usually, you are looking at an output of the last stage so that all previous stages are not loaded by the probe; there is not so much of a concern at the tail end of the IF. Yes, if you hang a scope probe directly on the grid or base of a VHF oscillator, you are likely to kill it.... but that's not something you would do in a normal alignment anyway.

Good Luck,

_________________
Jim T.
KB6GM


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Thu 10, 2020 1:06 am 
Member

Joined: Jun Sat 18, 2011 5:14 pm
Posts: 252
Location: Washington State 98226
Thanks dtvmcdonald. Your additional explanation helps and makes sense.

Thanks Jim for your response.

Quote:
If you want to observe what happens with an impedance mismatch, connect your scope probe to the tracking generator output. (I assume you are using a spectrum analyzer with a tracking generator


I assume by scope you mean the SA input. I don't own a spectrum analyzer yet but hope to purchase one soon (with a tracking generator). I want to make sure I know basically how to use one and how to protect it before I buy it.

I will either make a DC blocking device or get one like this.

Attachment:
DC blocker.pdf [1.55 MiB]
Downloaded 36 times


I assume there is just a series capacitor inside that sort of DC blocker. But no diode clippers.

John w.


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Thu 10, 2020 5:39 am 
Member
User avatar

Joined: Oct Sun 15, 2006 12:57 pm
Posts: 7440
Location: Liberty, Missouri
Gasman asked,
Quote:
...can be switched between 10X and 1X and if so is that because you don't want to accidentally switch it to the 1X position...
Precisely.

I don't bother with a blocking capacitor simply because it isn't needed. The possible current levels involved won't stress the SA's input circuitry.

For example;

The ratio of resistances in the 10X probe(9 MegOhms), and at the input to the spectrum analyzer(50 Ohms) is about 180,000 to 1.

So, If you were looking at a 20 Volt signal in a receiver, using your 10X probe, you would have a minuscule 110 micro-Volts at the input to your spectrum analyzer.

Now, if you were looking at that same signal at the plate of a tube having 350 Volts of DC on it, in addition to the 110 micro-Volt signal we now have, according to Ohm's law, about 0.4 milli-Amps of DC current. So now we have 110uVolts of signal, and 0.4 milli-amps of DC current, going into your Spectrum Analyzer. Ohms law says that the 0.4mA current translates to a Voltage of 0.02V or 20 milli-Volts, at the 50 Ohm input.

My ancient Spectrum analyzer's MAXIMUM INPUT limitations are;
Image


Which is FAR FAR FAR IN EXCESS of the signals we are going to apply to its input WORST CASE, therefore, no isolation cap or clamping diodes needed. I would think a modern digital Spectrum analyzer, would have similar input limitations.

_________________
Pics, Bold & underlined text, are usually links.

https://mikeharrison.smugmug.com/

Mike


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Thu 10, 2020 2:53 pm 
Member
User avatar

Joined: Nov Mon 02, 2009 7:01 am
Posts: 4812
Location: Lincoln City, OR 97367
Greetings to Mike and the Forum:

Good point Mike. However, the devil is in the details. Your calculation assumes that the scope probe is operating as designed and is good for the 350 volts of plate voltage. If, as many people do, we are talking about using ChiCom junk probes, that is open to question.

It also assumes that one will not be switching inputs to the spectrum analyzer to look at various things and will not inadvertently grab an input that exceeds the spectrum analyzer's ratings.

DC blocks are cheap insurance..... although one must be mindful of what they do to the unit's frequency response. For most of what I do, it is not important.

Sorry, I am a belt and suspenders guy. :D

Regards,

_________________
Jim T.
KB6GM


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Thu 10, 2020 3:36 pm 
Member
User avatar

Joined: Nov Mon 02, 2009 7:01 am
Posts: 4812
Location: Lincoln City, OR 97367
Greetings to John and the Forum:

As much as I like to wax eloquent, one picture is worth a thousand words, as they say. In this case, I have saved you from reading 3K words! :D

Here is a photo of my Rigol DSA-815 TG with a Tektronix 10X scope probe connected to its input and the probe tip looking at the tracking generator output. Note that the scope probe is very Hi-Z, while the tracking generator output is 50 ohms. Further, the probe certainly doesn't match the 50 ohm input of the analyzer, either. This is a full span (1.5 GHz) picture at 10 dB log per division. Not all that great, is it?

Attachment:
scope probe full span 10dB R.JPG
scope probe full span 10dB R.JPG [ 277.31 KiB | Viewed 618 times ]


Now here is the same setup with the same mismatches but with the center frequency set to 455 KHz and the span limited to 25 KHz. Also, the gain resolution has been increased so that we are now looking at 2 dB log per division. Looks pretty good to me.

Attachment:
scope probe 455 KHz 2 dB log R.JPG
scope probe 455 KHz 2 dB log R.JPG [ 300.28 KiB | Viewed 618 times ]


Lastly, here is the spectrum analyzer set to 10.7 MHz with 400 KHz span, also at 2 dB log per division. This also looks pretty good to me.

Attachment:
scope probe 10.7 MHz 2 db log R.JPG
scope probe 10.7 MHz 2 db log R.JPG [ 287.09 KiB | Viewed 618 times ]


As has been pointed out, absolute level isn't important in alignment; only the response curve shape is important. So, you can do good work with just a scope probe and don't worry about the impedance mismatch. The trick is in limiting the span that you are looking at. If you are going to be doing TV alignment, you might have to be more careful.... but aligning TV sets is best done with a sweep generator and conventional scope IMHO. Unless you are aligning one of the really old 26 MHz IF TV sets, get a B&K 415. The 415 makes TV alignment so simple that you will wonder how anyone could manage without one. Great instrument.

Regards,

_________________
Jim T.
KB6GM


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Thu 10, 2020 4:26 pm 
Member

Joined: Aug Sat 22, 2015 9:40 pm
Posts: 34
Location: Winchester, VA 22601
Good comments from all. My particular Siglent has rf input jack maximums of 50VDC and +30dBm, which for 50 ohms would be about 7 volts rf or 1 watt. Seems generous. I try and avoid maximums anywhere near those. I’ve seen HP’s with 0 VDC input on the front panel.

While acknowledging the 10X probe to 50 ohm input math, I sometimes use an outboard/inline dc block in the simple form of a .01 or .005 uF cap. Regarding “loading down the DUT” with impedance mismatches and tight or loose coupling from probes and resulting inaccuracies in measuring and alignment, I had a recent experience using my SA to align the IF’s of a single conversion Hammarlund. Using the mixer grid as the input of the SA TG and the detector as the RF output to the SA, I tried several various attachment methods. These included direct attachment to tube pins, attachment around insulated wires to tube pins, and variations of each using 1 meg and 10 meg resistors and a 10X Tek scope probe to the SA. In each case the peak amplitudes varied a LOT, as we would expect. However, in all cases the displayed patterns, bandwidth, symmetry, and even actual crystal resonant frequency remained the same. These examples were over a very limited +/- 10kHz scan bandwidth

Finally, my one inviolate Golden Rule regarding SA use is never, ever attach to a point, pin, or node in a DUT without first thoroughly characterizing the attachment point. Examine the schematic, know what’s supposed to be going on, and measure for any ACV and DCV values before proceeding. SA’s , with or without DC blocking caps, aren’t forgiving auto-ranging DMM’s and should never be treated as such.

Dan


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Thu 10, 2020 8:36 pm 
Member

Joined: Jun Sat 18, 2011 5:14 pm
Posts: 252
Location: Washington State 98226
Great info from all of you... I am feeling pretty confident that I could actually use a SA to do a visual alignment now --> without blowing up the front end! And the pictures are worth 3K in words. Thanks for taking the time to do that demonstration. :)

Will probably use a 10X probe and a DC blocking cap.

John w.


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Fri 11, 2020 3:12 am 
Member
User avatar

Joined: Oct Sun 15, 2006 12:57 pm
Posts: 7440
Location: Liberty, Missouri
Jim said;
Quote:
...It also assumes that one will not be switching inputs to the spectrum analyzer to look at various things and will not inadvertently grab an input that exceeds the spectrum analyzer's ratings...


Barring Probe failure it has been demonstrated there is NOTHING in a typical receiver that will exceed an analyzer's input limitations when using a 10X probe It's the "responsibility of the operator to know" the maximum ratings of the probes and any other test equipment he is using. The vast majority of TEK SCOPE PROBES will handle 300 volts RMS, or about 400+ Volts DC, and some up to 1.5KV. Import probes are, as you indicated, a mixed bag, but again they come with a spec sheet that should be read and understood.

Jim also said;
Quote:
...This is a full span (1.5 GHz) picture at 10 dB log per division. Not all that great, is it?...
One slight problem with pic #1. That probe is probably at best a 200MHz rated probe so anything above that is "no mans land", and, you have no ground on the probe tip (tracking generator) end which is a must for the frequencies you swept. However you get down into the HF and IF regions and it behaves quite well as you show.

I would be interested to see what your equipment might show for properly terminated 4MC wide slice at 26 and 45 MC.

WB4GRA said;
Quote:
...Finally, my one inviolate Golden Rule regarding SA use is never, ever attach to a point, pin, or node in a DUT without first thoroughly characterizing the attachment point. Examine the schematic, know what’s supposed to be going on, and measure for any ACV and DCV values before proceeding. SA’s , with or without DC blocking caps, aren’t forgiving auto-ranging DMM’s and should never be treated as such...


Good advise.

_________________
Pics, Bold & underlined text, are usually links.

https://mikeharrison.smugmug.com/

Mike


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Fri 11, 2020 11:21 pm 
Member
User avatar

Joined: Nov Mon 02, 2009 7:01 am
Posts: 4812
Location: Lincoln City, OR 97367
Greetings to Mike and the Forum:
Quote:
Barring Probe failure it has been demonstrated there is NOTHING in a typical receiver that will exceed an analyzer's input limitations when using a 10X probe


I don't recall anyone saying that a spectrum analyzer is limited to use on receivers. I use mine almost exclusively for transmitter or transceiver work. Not long ago, I had a scope probe on the grid of the PA in a 200 watt transmitter. True, I was using it to feed a scope and not my spectrum analyzer at the time, but imagine what might have been present there had the PA tube flashed over. 800 volts DC on the plate not to mention the RF component.

Quote:
One slight problem with pic #1. That probe is probably at best a 200MHz rated probe so anything above that is "no mans land",


My point exactly. This is not a viable use of the probe / analyzer combination. However, in spite of the probe's stated limitations, the actual performance, while not stellar, is not too bad... I believe what the instrument is trying to tell me.

Quote:
you have no ground on the probe tip (tracking generator) end which is a must for the frequencies you swept.


Obviously not, since the display is there. In this case, the two type N connectors on the spectrum analyzer are connected together in the same RF block with a shorter distance than the length of my ground lead, so no ground lead is necessary. Had I connected it, the inductance presented would render it invisible anyway. The whole point of the picture was to prove that you CAN'T do things that way, but you CAN do things if you limit the sweep span.

Of course, when probing a separate piece of gear, the ground lead is necessary.

However, if one intends to demonstrate the difference between an unrealistic use of a scope probe and spectrum analyzer versus a realistic one, I think the photo series accomplishes exactly that.

Regards,

_________________
Jim T.
KB6GM


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Sat 12, 2020 5:29 am 
Member
User avatar

Joined: Oct Sun 15, 2006 12:57 pm
Posts: 7440
Location: Liberty, Missouri
Jim said,

Quote:
...I don't recall anyone saying that a spectrum analyzer is limited to use on receivers... I use mine almost exclusively for transmitter or transceiver work...This is not a viable use of the probe / analyzer combination......However, in spite of the probe's stated limitations, the actual performance, is not too bad... I believe what its trying to tell me...
I agree with you regarding transceiver/transmitter use, I do the same.

However, the OP's opening statement/question was specifically about receiver I.F.s which is what I was addressing, nothing more. No DC to blue light broad band response, No 800 Volt Transmitter PAs, No "what if's", No Highly accurate voltage or power measurements, I simply proposed a method by which one can easily evaluate a receiver's I.F. filter response with a minimum of loading, or possible damage to equipment.

_________________
Pics, Bold & underlined text, are usually links.

https://mikeharrison.smugmug.com/

Mike


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Sat 12, 2020 3:14 pm 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 22213
Location: Somers, CT
I've seen too many warnings about relying on a DC block to protect an analyzer on the HP Forums to fully trust one to protect an unreplaceable mixer in my Tek 497P. The charging current on the cap when hitting a B+ plus line accidently can blow the diodes. I'd want a -20dB power attenuator with a DC block in front of it. For most sweep ranges adding a cheap MCL RF transformer inline will also limit maximum RF passthrough since the tiny cores will saturate to limit RF.

For sweeping a receiver an analyzer and tracking generator are really overkill and present a few obstacles as well--I've used my 497P and its companion tracking gen to do sweep alignments but I much prefer using the HP-3336B along with a cheap Tek XY display and a cheap log amplifier/detector board from China via eBay. This avoids having to have the detector tracking the generator; and the generator can sweeping at the RX operating frequency while the log detector is sampling the IF bandpass. I ususlly inject at the mixer grid to avoid loading the first IF, the mixer acts as buffer.

_________________
Just because it can be done doesn't mean it should be done.


Top
 Profile  
 
 Post subject: Re: Using a Spectum Analyzer to do a visual alignment
PostPosted: Dec Sat 12, 2020 7:06 pm 
Member
User avatar

Joined: Oct Thu 04, 2018 2:11 pm
Posts: 489
Location: Suburban Chicago
We use spectrum analyzers all the time for both RX and TX work at my job as a custom radio designer. We use direct 50 Ohm connections to the TX output through appropriate attenuators to do TX power and signal quality measurements. For everything else we us an ungrounded FET probe. In our case everything else is diagnosing RX/TX issues on new designs and production units that fail the automated testing. Basically the FET probe is being used as a glorified signal tracer so the fact that it is ungrounded is of no consequence. The probe must have come with a ground clip of some kind and someone lost that for us years ago. We rarely miss it and one of our probes may still have that accessory if the need arises. Tracking generators mostly went out of style and were replaced by network analyzers quite some time ago. We use network analyzers to sweep both RX and TX paths. We rarely have to worry about mixers, these days the in band RX and TX signals disappear into and exit from some monstrous custom IC that more often than not exchanges bits with the rest of the radio rather than analog IF signals. Life is a lot different these days than it was when I started in 1974. Spectrum analyzers are a useful tool for radios both new and old, however many of the older radios people are working with here operate in frequency bands that are well within the capabilities of modern oscilloscopes so the need for an SA here is limited. If you happen to have one you will certainly find it useful, just make sure to protect the input. That is a constant struggle for even seasoned RF designers, especially when they return from lunch and to their horror discover that the software engineers are gathered around the RF bench and are trying to hook up and measure a radio....


Top
 Profile  
 
Post New Topic Post Reply  [ 27 posts ]  Go to page 1, 2  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: jnovotney, khutch, KX5JSC and 31 guests



Search for:
Jump to:  


































Privacy Policy :: Powered by phpBB