PDM Generator help

[K9ACT]

First of all, am I correct in presuming that we can not post or attach images to these postings?

I removed both transistors and hooked the scope to the common bases.

http://schmidling.com/CARRIER.JPG is with the carrier level set to 1 cm and no audio.

http://schmidling.com/100PERC.JPG is with what a believe would be 100% modulation at 1kc.

It seems to be doing what it is supposed to but I don't know what all the fur is.

The counter says the freq is 172kc and I thought it was supposed to be around 100kc.

If I put an earphone with 50 ohm in series across the test point I hear audio doing what it should.

How am I doing?

js

[kf1z]

Right, you have to host the pics somewhere else, then link to them from here, and then they'll show up in your post.

Your pics look right, though I presume you're using a fairly slow scope? 20 maybe 50mhz?

172khz is closer to being correct. I think 160khz is the target...

100khz too low.

My old one was 210khz with the caps I bought (Rev C board, no kit of parts)
They were 10% or more tolerance, so I changed them out and got it down to 180kc or so.

The "Fur" is the modulation.
You are seeing the trailing edge moving with the audio... this is a good thing!

I would guess that since the modulation is all in the positive direction, that the negative peak limiter pot is set nearly fully counter-clockwise.
The trailing edge should be closer to the leading edge at near 100% negative modulation.
But that is an adjustment for later.

The main thing is that it is working.

[K9ACT]

Removing the load, (50 ohm resistor and headphone) squared it up nicely.

Now I can see what the controls do but I need some help understand why and what they mean.

The carrier level changes the duty cycle but I have no idea how this relates to the RF world downstream or what determines where to set it.

The audio level increases the duty cycle starting where the carrier ends. It would seem like 100% modulation would be when the total is twice the carrier width but as this can be accomplished over a wide range of carrier levels, I don't get it.

The NPL is still more mysterious. I can move the point where the modulation begins relative the the edge of the carrier but I don't understand what this accomplishes or how.

I am trying to relate this to my CBS Volumax and I am lost.

And yes, it's a 20 mhz scope.

js

[Gilly]

Hi Jack

This is one of the photos I took when building my first PWM....



To be honest I had know idea what I was looking at either.....
But when the Modulator was finished and tested into a dummy load say 5 - 10 Ohms or so...looking at the analog waveform and going back to the PW signal you are able to make some sense of what its all about...
Varying the input analog signal frequency / amplitude and adjusting the Negative limiter etc: and just general playing around....
I have built 3 PWM now and they all work so well....

Post more pictures as you go .....


Wayne

VK3ALK

[K9ACT]

Thanks for the encouragement but how did you post that pic?

It's a real pain to put them on my website if I don't have to.

js

[Gilly]

Hi Jack

My friend gave me some room on his WEB server and uploading is done with a simple explorer type program that appears on the screen....
Don't worry about understanding whats going on......in fact forget about it !!! .
The main thing is what Bruce said...its all working so thats good....
Finish the modulator connect a dummy load ... check the 12 volt digital signal on the gates and apply a low test voltage to the drains and I sure it will all work fine.....
The carrier adjust control will all make sense then etc:

Wayne

[kf1z]

Removing the load, (50 ohm resistor and headphone) squared it up nicely.

Now I can see what the controls do but I need some help understand why and what they mean.

The carrier level changes the duty cycle but I have no idea how this relates to the RF world downstream or what determines where to set it.

The audio level increases the duty cycle starting where the carrier ends. It would seem like 100% modulation would be when the total is twice the carrier width but as this can be accomplished over a wide range of carrier levels, I don't get it.

The NPL is still more mysterious. I can move the point where the modulation begins relative the the edge of the carrier but I don't understand what this accomplishes or how.

I am trying to relate this to my CBS Volumax and I am lost.

And yes, it's a 20 mhz scope.

js

This is why it's called a Pulse Width Modulator....

The width of the pulse determines the DC voltage the modulator produces at the output of the filter.
The wider the pulse, the more DC voltage.
What determines where to set it (the carrier level)will depend on the RF deck design, it's impedance... and the modulator power supply voltage.
The modulation should add AND subtract from the width of the pulse... so if it only ADDS to the carrier level, then the NPL is set to low (or high?) that's why I made the comment...
The pulse should be almost non-existant at the peak of negative modulation.. (near 0 volts output)
In the picture of the scope trace, this would mean the "fur" would extend from the leading edge (left side) of the pulse to a point about the same distance to the right of the point where the carrier width of the pulse was. Assuming a audio generator is used to provide modulation...
Voice characteristics (assymetry) will be a different story...

A good place to start will be to adjust the pulse-width to approximately 1/3 of it's total\, or less.

Then later, with the RF deck up and running, you can adjust the carrier level to it's designed level.

[K9ACT]

Hi Jack

My friend gave me some room on his WEB server and uploading is done with a simple explorer type program that appears on the screen....
Wayne

But how did the picture get embedded in your posting if you put it on another web site?

js

[kf1z]

Jack,

Above the box where you type you post, there are some buttons.
One of them is "Img"

(img)(/img)
It'll look like that in the message except the () will be [ ] instead
Type or paste the entire Url where the picture is stored online, including http:// in between the ] [

Like this

(img)http://www.mypics.com/photo1.jpg(/img)

again, it'll be [ ] instead of ( ) ... those are here just so you can see the tesxt that will appear when you are composing the message.

Then when your post is displayed, the picture will be retrieved and displayed.
The picture must reside on another server.

BRuce

[K9ACT]

This is an update of scope views of my board looking at the output of the IC as I am waiting for the replacement buffer transistors.



Let me know if my understanding this is correct. Modulation is a 1kc sine wave.

Jack

[steve_qix]

Hi Jack,

The PWM looks perfect. Usually, the "carrier" duty cycle is set to between 30% and 40%, and 100% positive modulation takes place at twice the carrier duty cycle. This leaves headroom for positive peaks in excess of 100%. I usually run my rig at about 33% for carrier, allowing upwards of 200% positive modulation.

But, what you have there looks quite normal for a 50% no-modulation duty cycle.

Regards,

Steve