Alignment Tools?

[JBL GUY]

Total Harmonic Distortion (THD) is the ratio of the harmonic power to the fundamental power. It is computed by searching the entire spectrum to find the peak frequency (fundamental) and then calculating the total power in the harmonic frequencies.

The THD level is then computed as the ratio of the total harmonic power to the fundamental power. Residual noise is not included in this calculation, although when one measures THD+N (N being the noise component) it is included.

Short answer, without an input (the fundamental frequency) the software is unable to calculate the amount of distortion.

Here is some additional information.

An FFT size of 2048 or greater is recommended in order to provide adequate spectral resolution.

Hanning or Blackman smoothing windows are recommended because of their low noise quality.

 

[super98lsc]

Ok I got the software sorted out. Even though it was measuring my external feeds on the screen/scope windows. I had to switch the Source selection for the THD from internal to external. At that point the distortion % displayed came much more in line with something I would expect to see. I was using Signal Generator app on my Iphone 5S input to the line-in jack on the PC/Precision workstation.

Now at 1Khz sine it measures .073% THD and .1113% at 440HZ

 

[rcs16]

Hi guys,

Measuring the THD of the modulation signal is only one source of non-linearity. The modulation of the carrier adds more distortion, how much is it and how to measure it ?
I guess i have to look at the 8640 Service manual to see how they measure the THD, it is so long ago. I know there was no 8901A at the time the 8640 was designed.
To measure the THD on the 8656B, I use as documented, a calibrated 8901A and a Amber 3501 which would sub for a 339A or similar THD analyzer.
Adding the 8901A into the signal path adds to the THD. Even if I use the Amber as the modulation source,>100x less THD than the osc in the 8656B. The results do not change much, as the modulation and demod of the 8656B and the 8901A are significantly higher.

 

[super98lsc]

Well I missed the end of the auction on the Leader unit I was after. I have done some more reading and seems the HP's are held in higher regard? I cannot tell from the specs but does the HP 8656B do FM-Stereo? Or do I need to buy another modulator to handle that portion? Sorry for the 2000 question I just want to buy a quality unit that will last and provide the most use for years to come. I have a pile of Sansui's I am rebuilding. (no plans to sell them though)

 

[rcs16]

8656B does not have any stereo or RDBS modulation capabilities. You will need a external unit for that. Most tuner alignments only need mono FM modulation. Some higher end tuners have adjustments for stereo separation and other things like pilot cancellation.
8656B is obsolete!! If it dies, you are responsible to fix it.
If you want something new, like a Rhode& Schwartz unit you will pay way more, but it will be serviceable/calibrated by the mfg, at a great cost.
I have no idea if Panasonic still makes, supports their RF SG's.

Good luck
Rick

 

[ConradH]

Updated my post #39 with numbers for the Sencore SG-165 and Leader LSG-231.

 

[super98lsc]

Thanks Conrad, I am hitting a local surplus private Auction deal Monday and will see what I can find. My IG-37 showed up last night. Its in MINT condition. All the pots work smooth, I hit them with some De-Oxit anyhow, it shipped with the test leads and cables too. I tested all the capacitors (paper and the big multi-cap can) and they are all still well within spec. I will fire it up here soon on the Variac and ensure no BBQ action comes into play. Someone took care of it and built it nicely.

 

[rBuckner]

Wow, that is a clean unit! You're on your own with the RF frequency it looks like. You can get it close and then fine tune the tuner, no different really than an SG-165. Have fun playing with it! :yes:

 

[rBuckner]

Apologies for the following (minor) thread crash!

I did land a Panasonic VP-8131A on Friday. I also got two PDF's from another eBay vendor who has a parts unit that together make up the operating/service manual. I'm not sure about grabbing the parts unit yet but will start another thread when I get the "good" one in.

 

[rcs16]

Nice Heathkit RF SG, hope it does the job for you.

Hey rBuckner,
Very nice specs, for the Panasonic VP-8131A . Way better distortion than the HP 8656B and many others. What do plan on using the unit for? I assume broadcast tuner repair, alignment! What were you using before? Good to see you got the stereo encoder version too.
Not sure if these are still in production, supported by Panasonic, with spare parts avail, etc, if not I'd suggest to get a service manual and a spares unit if you want to keep it a long time. I will look forward to your Panasonic VP-8131A thread.

 

[ConradH]

Nice looking Heathkit! I assume the bulb is for the oscillator, so it works like the old HP signal generators. That's a good thing. The reason the SG-165 is so bad is they start with a square/triangle generator and convert to a sine wave using a somewhat questionable method, at least compared to how people do it today. My Wavetek function generator does the same thing, but gets well below 1%.

FWIW, there have been great debates, probably in the Yahoo FM group, about the significance of generator THD. The fear is that the tuner can be adjusted to cancel out the generator distortion, so you think you're minimizing it, but end up being a ways off. Imagine the transfer function of the generator is curved a bit one way, and the tuner gets aligned so it curves the other way. I can see how it would happen in theory, but I don't know how big an issue it is in real life. I've used the SG-165 a few times for THD and sound-wise, the alignment was great, in spite of its high THD.

It also seems like an easy mod to add an LC tank to the generator output to kill most of the harmonics and greatly improve the situation, but as mentioned above, goodness of the modulation is anybody's guess. No idea how to measure that, other than having a really good tuner that's already correctly aligned. Chicken and egg.

 

[rcs16]

Hi ConradH,

You seem to be getting your THD specs a bit mixed up. Adding a L/C tank to the SG o/p will reduce THD and spurious signals of the modulated carriers RF. But it will do nothing about reducing the modulating source THD/noise and the linearity of the freq modulator itself. I could actually increase the modulated carriers detected THD, because if you reduce the sidebands, that could create more THD. Does any of this make any sense? I certainly am not a Bezel function expert, that is for sure.
Yes you are correct, about chicken and egg, usually you want the measurement instrument to be >10x better than what you are measuring.
Measuring modulated FM is a science in itself. I know for the HP 8656B, the performance verification test calls for a HP 8901A. The 8901A is verified using a special test jig, HP 11715A.
So the 8901A would be great to use with RF SG with >=0.1% modulated THD.
I would like to read the service manual for a Panasonic VP-81xx series to see what they use to verify its modulated THD performance. I am guessing they use a RF spectrum analyzer as a detector and a fft or THD analyzer to measure the harmonics.

 

[super98lsc]

Well I have one 2000x that is really strong and sounds clean which I use for a baseline of what good sounds like at least to me. I connected the marker output to the o scope and with my scopes freq counter I was able to at least validate the 400hz and 1kHz. It does generate on the others but I haven't connected it to my PC yet as a freq counter or checked its THD yet. Just went to a local electronics supply to snag a chassis mount female BNC connector to replace the RF output and an RG 58 BNC test lead with the ez grips.

 

[ConradH]

Hi ConradH,

You seem to be getting your THD specs a bit mixed up. Adding a L/C tank to the SG o/p will reduce THD and spurious signals of the modulated carriers RF. But it will do nothing about reducing the modulating source THD/noise and the linearity of the freq modulator itself. ....

Not mixed up, or not too much I hope, but I could probably pick my words better. An LC tank would always be beneficial after whatever audio oscillator circuit is used, and the output of that is the only thing I can measure for THD. Everything after that also contributes, but there isn't much I can do about it, nor can one just slap a filter on it! In talking about the cancellation of distortion, I'm thinking of when you "warp" the curve to minimize distortion during alignment, that it may be possible to cancel some of the total distortion from the AF+RF+MPX combination, resulting in misalignment. IOW, the minimum THD reading, whatever it is, will be at a different slug position for a low distortion generator system vs. a high distortion one. Or at least that's what I got from the Yahoo FM group.

FWIW, I'd guess one could get surprisingly close to a good THD adjustment by ear, with a good station and a bit of practice. I haven't tried it, but I was amazed at the sonic difference between my Yamaha CR-620 being a little out vs. well aligned. I also have a 2000x and it's a great unit to learn on, having 6 slugs to mess with in the IF, errors in the procedure and great performance when done. IMO, Sansui was really good with RF.

 

[rBuckner]

Very nice specs, for the Panasonic VP-8131A . Way better distortion than the HP 8656B and many others. What do plan on using the unit for? I assume broadcast tuner repair, alignment! What were you using before? Good to see you got the stereo encoder version too.
Not sure if these are still in production, supported by Panasonic, with spare parts avail, etc, if not I'd suggest to get a service manual and a spares unit if you want to keep it a long time. I will look forward to your Panasonic VP-8131A thread.

I rebuild a lot of solid state receivers for folks, mostly Marantz. I use an SG-165 now and as Conrad has experienced, it does do a good job for me. I'm just picky and wanted something really nice. Also if the SG-165 goes kaboom on me I didn't want to be RF-less! I mentioned before that I put a BNC jack on the front of my SG to inject clean audio from my HP 8903B analyzer.

Back to the Panny, they don't make it any more but sold it off to a company named Levear. Those aren't distributed in the USA that I can find but look exactly the same. I mentioned that I have the SM in PDF form and took a quick look at the schematic before pulling the trigger. I also look forward to the evaluation thread!! I also have a line on a spares unit but need to work the price down some more.

 

[super98lsc]

Do I need a VTVM? Appears that many alignment procedures call for one including my Sansui. Says they higher impedance to prevent loading down the circuit like a digital or solid state VM would.

 

[ConradH]

I don't know what the input impedance of VTVMs was, but they probably had an input resistor of some sort, likely no greater than 10 megohms. Many DVMs (my ancient Fluke 77) are 10 megs on the input. Nice meters like the Agilent 6.5 digit meters will have a setting to make the input impedance almost infinite. They have to, or they'd load the circuit under test so much that their accuracy would be wasted. It's the mechanical meters that were a big problem, with inputs rated in ohms per volt. I think a Simpson 260 was 20 kohms/volt.

 

[super98lsc]

Ok so my fluke 87 true RMS which is about 12-14 years old now I think should do this function fine then. I was trying to figure out how on earth an old mechanical tube meter could superior to a modern high end one. Not knocking on the old stuff but seems they would not "lose" functionally in 50 years of R&D.

Have a few beginner questions...

When the manual says connect o-scope to Output Load. Which output are they referencing?

Also the IG-37 RF output is centered on 100hz. Shouldn't it "broadcast" on other frequencies as well? Example in the sansui instructions they want you to tune the dial to 90/106 and some other spots for different steps. I'm thinking I need to go through the IG again and be certain someone didn't remove or setup something inside into a calibration mode and left it there electrically speaking.

 

[gadget73]

Most VTVM's are 10 or 11 megohm input. VOM's of the time might have been 20k ohms / volt. The big advantage to a VTVM if you need to set something to zero volts is that you can center the meter and more easily see it. I also find them easier to use for doing peaks because you can see where the needle peaks more easily than where the scrolling numbers top out. Electrically both should do the job but I just find the VTVM is more user friendly when you don't actually need to hit a specific voltage. I keep one on the bench and its mighty useful for certain things. I use it mostly for radio work because my digital displays in ancient Egyptian around RF. The VTVM just gives me voltages without any flak.



I'd guess output load is your load resistors on the speaker output terminals, but thats just a guess.

 

[rBuckner]

...Ok so my fluke 87 true RMS which is about 12-14 years old now I think should do this function fine then.

When the manual says connect o-scope to Output Load. Which output are they referencing?

Also the IG-37 RF output is centered on 100hz. Shouldn't it "broadcast" on other frequencies as well? ...

I also think your Fluke should be fine, I don't own a VTVM.

I think they mean the amp output too but I usually monitor the tape out if it's needed.

I thought the knob on the lower right of the IG-37 varied the RF's frequency. I had commented that there was no marking for what frequency you were at earlier. Try that.