My Eico 720 project

[w1vtp]

I finally worked up the nerve to post my Eico 720 project.  The idea behind this project was to use off the shelf parts to put a nice sounding AM signal on the air.  I already have a pro-grade Bogen PA 60 watt amp so I used that for the audio.  The mod iron was a Hammond P-T1650P push-pull hi fi output trannie connected backwards.  The mod reactor is a Hammond P-T193C 20 hy @ 100 ma choke.  The choke has a 600 V rating  so I plan on not grounding the case.  I'll protect the case from accidental contact.   

A modern solid state amp would have resulted in better performance because of lower source Z.

I think I got some pretty nice results -- check out the PDF file.  Freq resp is decent and I think the THD results are acceptable.  The last frame in the PDF has nothing to do with the experiment except that I plan on inserting it into the audio chain so that I can control the 720 freq response.

Thanks to Paul WB2SKC for his help.

73, Al

[k4kyv]

A modern solid state amp would have resulted in better performance because of lower source Z.

That shouldn't make any  difference, since in plate modulator service the amplifier is working into a constant load.

If you were using the amplifier to drive a higher power class B or class AB2 stage, then you would need the lower source Z.

[AB2EZ]

Don

I respectfully disagree.

If the effective source impedance of the amplifier (i.e., as measured at the output of the step-up transformer) is significantly lower the the modulation resistance presented by the rf tube(s)... then the frequency response will be extend further at both ends of the audio response passband.

At higher frequencies, when the reactance of the r.f. bypass capacitors becomes comparable to, or perhaps even lower than the modulation resistance... the amplitude of the modulation of the plate and screen voltage will not roll off as fast. I.e. the modulation source will deliver the extra current without a significant drop in its output voltage.

Likewise, at lower frequencies, when the reactance of the Heising choke... in parallel with the reactance associated with the magnetizing inductance of the transformer... falls below the modulation resistance, the modulation will not roll off as quickly if the effective modulating source impedance is significantly lower than either one of these.

The price you pay for this is that the audio amplifier has to have a higher power rating, in order to drive power into an effective load that is higher than its output impedance... but high power audio amplifiers are readily available these days if you are using a solid state model.

Stu

[W3RSW]

Reminds me of the damping factor wars for speakers.

Impedance of driving amp had to be 'way' lower than the actual load or somesuch. 

All that kind of went away with op. amp. type circuitry.

Progress.

[AB2EZ]

"Yes, but what if you never use any r.f. bypass capacitors like many of us don't and rely more on properly designed audio sections and final modulators that don't over-modulate the RF amp?"

I'm not sure I understand your comment. Perhaps my prior post was confusing

I'm not talking about a low pass filter used to compensate for over modulation or non-linearity of the modulator (sometimes referred to as a splatter filter).

If you have a dc blocking capacitor between your rf output tube and your output network (e.g., typically .001 uF or larger), and a safety choke or a link-coupled interface to your feed line ... then you have the following path: modulated B+ =>  rf choke => tank coil => safety choke => ground, which represents a 27k ohm path from the modulated B+ to ground at 6kHz.

If you have a .001uF (or larger) bypass capacitor to ground at the modulated B+ side of your rf choke ... then you have another 27k ohm  path from the modulated B+ to ground at 6 kHz.

Those capacitive paths to ground draw current from the modulating source.

Likewise for the screen r.f. bypass capacitor if you are using a tetrode or a pentode.


Stu

 

[Steve - WB3HUZ]

Nice work! The numbers are quite acceptable. How did you measure THD?

I've done something similar with my Eico 720 in the past. I used a UTC LS series hi-fi output tranny and a 400 watt solid state audio amp. Needless to say, I had lots of audio. I didn't make good measurements like you did though.

Hope to hear the setup on the air sometime soon.

I finally worked up the nerve to post my Eico 720 project.  The idea behind this project was to use off the shelf parts to put a nice sounding AM signal on the air.  I already have a pro-grade Bogen PA 60 watt amp so I used that for the audio.  The mod iron was a Hammond P-T1650P push-pull hi fi output trannie connected backwards.  The mod reactor is a Hammond P-T193C 20 hy @ 100 ma choke.  The choke has a 600 V rating  so I plan on not grounding the case.  I'll protect the case from accidental contact.  

A modern solid state amp would have resulted in better performance because of lower source Z.

I think I got some pretty nice results -- check out the PDF file.  Freq resp is decent and I think the THD results are acceptable.  The last frame in the PDF has nothing to do with the experiment except that I plan on inserting it into the audio chain so that I can control the 720 freq response.

Thanks to Paul WB2SKC for his help.

73, Al

[w1vtp]

Steve

THD was taken by measuring the harmonics of the modulated component on a spectrum analyzer.  Go to TX and adjust the sideband of the fundamental (eg @ 1KC the first sideband is 1 KC from the carrier) so that it is ~6 dB down from the carrier -- that's 100%.  Then set the reference so that the harmonics (2KC, 3KC etc) can be jotted down.  In my case I had generated an Excel spreadsheet that allowed inputting the #'s directly into the spreadsheet.  The dBs are then calculated into %.  The math is "root of the sum of the squares."  So I just did up to the 5th harmonic.  The spreadsheet then automatically calculates the THD for that frequency.  I had separate worksheets for 50, 100, 500, 1000, 2000 up to 6 KC.  Then a graph on a separate worksheet looks at the results on each sheet and draws the curve.

I gotta say that 'lil 720 did yeoman service with all those "buzzard transmissions" while Paul and I were taking the measurements. We wiped out a new GZ34 doing the experiment -- I'm going solid state in the rectifier from now on.

If anyone wants the spreadsheet I can upload it.

Thanks for the encouragement, 73, Al - Violet Toilet Paper - W1VTP, that is

[Steve - WB3HUZ]

Cool. I figured you did it this way when I saw the pix of the spec an.

Please upload or email me the spread sheet. I'll make it available to all, if you wish. I made similar measurements on my FT-102 with the homebrew NE602 modulator, but I was too lazy to calculate the THD.  The spreadsheet would have been handy.



http://amfone.net/Amforum/index.php?topic=4888.msg112972#msg112972

[w1vtp]

Cool. I figured you did it this way when I saw the pix of the spec an.

...

This here board does not allow such stuff. 

Here's a link.  Input in the yellow fields.  Careful - some of the other fields have hidden 'figgerin" in 'em.  The graph should update automatically as you enter data

http://homepage.fcgnetworks.net/w1vtp/THD/THD_CALCULATOR_AND_PLOTTER.xls
Al VTP

[Steve - WB3HUZ]

Very nice. TNX.