How should the 'pulse train' look?

[dalbert02]

I am just learning about how pulses are generated by our transmitters and how they are then modulated onto a carrier by a pos or neg shift in frequency. I wanted to see what one pulse would look like, I assumed a simple square wave. However, when I used my FMA servo testor, things looked a little different. Is this how the wave should look? Thanks,
-dave

[kschauwe]

Is the input to the scope set to AC or DC coupling? Should be set to DC.

[jeffs555]

Even set to AC coupling the signal should still look like a rectangular pulse. That looks like it is coupled through a small capacitor into a low impedance load. Possibly a bad probe, or maybe just using a 1 meg probe with the scope input set to 50 ohms.

[MatC]

No, it shouldn't look like that. It should look square. Scope AC inputs normally have a time constant of ~0.5s so that's not it. Did you remember to connect the earth on the scope to negative on the servo tester? :)

[Daniel7866]

I am just learning about how pulses are generated by our transmitters and how they are then modulated onto a carrier by a pos or neg shift in frequency. I wanted to see what one pulse would look like, I assumed a simple square wave. However, when I used my FMA servo testor, things looked a little different. Is this how the wave should look? Thanks,
-dave

You sould get a rectangular signal as servo pulse repeated about every 20msec (not critical).
Within those 20msec you can "fit" several servo pulses to control different servos providing that you left a small "rest time" to separate pulses and a longer rest time to tell the Rx the begining of a new frame.
Go to http://www.datasheet4u.com/ and look for two old circuits: NE5044 & NE5045. Good info there.
Also read http://homepages.tesco.net/ada.tippett/Radio3.htm .

Daniel

[dalbert02]

Thanks guys! My ground was mechanically connected but not electrically connected to the circuit. Once I secured it properly, it became a normal square pule.
-dave

[MatC]

What do I win? ;)

[Malc C]

Not sure if its now relevant as you seem to of resolved the issue, but this is the PPM pulse train from my JR X3810 trainer port

http://www.micro-heli.co.uk/scope.jpg

[dalbert02]

Thank you Malc!
If you are ever in Miami, the beers are on me!
-dave

[Malc C]

Thank you Malc!
If you are ever in Miami, the beers are on me!
-dave

I'll hold you to that... but on my salary it may be a long time before I can take you upon the offer !

[dalbert02]

Hey that goes for anyone on this thread! MatC wins the first cold one!
Here are some pics (sorry a little blury, my camera doesn't focus on the beam too good).

Probably nothing new to you, but maybe of interest to other newbies like me.

Thanks again,

-dave

[MatC]

That looks healthier, but those pulses don't look very square? The pulse train coming out of my tx looks much more like Malc's screenshot.
There's no overshoot after each pulse, so it's not AC coupling this time, it could well be that's just what it outputs.
So you've got the hang of PPM now? Pulse length is measured edge-to-edge, the low pulses are of constant width (300us), there is one extra long-pulse to sync things, and normal pulses are 1 to 2ms long.

[dalbert02]

I think I got it. However, I expected the pulses to be more square too. Also, my pulses seem hi and go low between each pulse. I thought it would be low and go hi. My scope was set to DC and ground was connected to the Tx's ground (battery negative) and the probe was clipped onto PPM out. Do my pulses still look a little wierd to you?
-dave

[village_idiot]

It is probably a condition where the line is pulled up to the operating voltage, and the chip grounds the line to make the pulse. (for some reason I can't remember the proper name for this????)

[village_idiot]

Also are your scope probes properly frequency compensated? You could have some low frequency issues with the probes, if your scope has a test pulse, you might want to try calibrating the probe.

[dalbert02]

My scope has no test pulse, just a test 1v and 10v DC. It is an old HP. I need a function generator. Any suggestions on a relatively inexpensive one?
-dave

[Daniel7866]

I think I got it. However, I expected the pulses to be more square too. Also, my pulses seem hi and go low between each pulse. I thought it would be low and go hi. My scope was set to DC and ground was connected to the Tx's ground (battery negative) and the probe was clipped onto PPM out. Do my pulses still look a little wierd to you?
-dave

The pulses from the trainer port are well from high to low. See pic from my FF9(9C) Futaba.
Since your pulses have some decay I hink that there is still AC coupling somewhere. Maybe inside Tx. What brand/model do you have?

Daniel

[Acetronics]

I think I got it. However, I expected the pulses to be more square too. -dave

Hi,

Noooo, Pulses ( at the coder output ) have a limited slew-rate not to generate "overshoot" in the freq. sweep of the HF Head ... ;)

Alain

[dalbert02]

My Tx is a Futaba 9CAP. I measured the PPM lead without the Rf module attached. So Ace, I am good then? I am seeing what I am supposed to be seeing? I kind like Daniels. They look nice and square.
Thanks,
-dave

[Acetronics]

Difficult to say ...

the slew-rate can be limited in the coder's part or at the HF Head input ... depends on the Tx model.

see here LM3900/IV section ... for example.

http://home.nordnet.fr/~fthobois/TF7N.htm

Alain

[Daniel7866]

My Tx is a Futaba 9CAP. I measured the PPM lead without the Rf module attached. So Ace, I am good then? I am seeing what I am supposed to be seeing? I kind like Daniels. They look nice and square.
Thanks,
-dave

Dave,

The pulses on my previous pic are from pin #3 of the trainer port. Ground is pin #2. That's the square 6 pins at rear of Tx.
___-___
l 3 2 1 l
l 6 5 4 l



The following pics are from pin 1 HF of module location:
39 & 40: HF module in place, no Xtal.
43: HF module removed.
In addition to the "inverted" signals between pics, you can see a amplitude difference too.
I don't know why.

Daniel

[andrewm1973]

It is probably a condition where the line is pulled up to the operating voltage, and the chip grounds the line to make the pulse. (for some reason I can't remember the proper name for this????)

The proper name is Open Collector. However I DON'T think this is open collector anyways. The leading edges look more intentional.

[dalbert02]

Thank you. I will try from the trainer port as well. Right now I am connecting to the pins the module plugs into. More pics soon...
-dave

[MatC]

I don't think it's an ac connecting issue, as then the pulses would show show "droop" on one side of the edge (as seen) and then over-correction on the other side (not seen).

I can only think it is intentional for whatever reason (something to do with RF slew rate control at a wild guess, as mentioned).