Jump to content
Sign in to follow this  
cyber-flyer

Sub-topic: Modulation of GPS serial interface

Recommended Posts

Thanks for the link Mr. RC-CAM. It's interesting but the device is still too big.

I believe the whole modem can be part of PIC or ATmel chip with the rest of the chip doing conversion from GPS to AX.25 format. In addition I will need 10 analog inputs with two of the inputs having 16 bit resolution. Basically I am looking for MIM-2 chip supercharged to 9600 bps and 20 Hz refresh rate. I am also restricted to FSK modulation so that AGWPE driver can decode the signal. So you know all my specs now :)

Regards,

cyber-flyer

Share this post


Link to post
Share on other sites

Yup, I agree. 9600 baud FSK on a PIC or ATMEL is not a problem. What is the 16 bit ADC inputs used for? Are they the Pitot and altimeter sensor inputs?

FWIW, with the typical noise floors on a micro equipped with A/D, the usable resolution is probably no better than 12 bits. More than that would take one heck of a clean design.

Share this post


Link to post
Share on other sites
Yup, I agree. 9600 baud FSK on a PIC or ATMEL is not a problem. What is the 16 bit ADC inputs used for? Are they the Pitot and altimeter sensor inputs?

Sorry Mr. RC-CAM I am bit delayed with my answers. Yes, 16 bit inputs would be used for pitot and pressure sensor measuremnts. I agree it will be tough to get full 16 bit resolution but it's better to oversample than undersample, do you agree?

I haven't pushed the project any further than this theoretical discussion due to lack of time on my side :( . What would be a benefit to use 12 bit rather than 16 bit converters (besides incremental cost difference)?

Regards,

Val.

Share this post


Link to post
Share on other sites
What would be a benefit to use 12 bit rather than 16 bit converters (besides incremental cost difference)?

In some cases, unless you design a 4-layer PC board using a careful layout, and/or use a tremendous amount of averaging, the 16-bits will best be truncated to about 12-14 bits anyways.

Even 12-bits is hard to maintain on home brew A/D based circuits. At 1mV per bit, the noise goblins are hard to tame. I am battle worn from those sort of ordeals.

So, although there are exceptions, there is often no advantage to the extra bits in a typical home-brew A/D equipped project. Except for bragging rights ;). But either way you are covered.

But to answer your question: The advantage with 12-bits is that you can pack the transmitted telemetry data (byte + nibble method) and reduced your bandwidth needs. That is to say, two sensors would only require three bytes to be sent (versus four bytes if you used 16 bit data).

Share this post


Link to post
Share on other sites
Even 12-bits is hard to maintain on home brew A/D based circuits. At 1mV per bit, the noise goblins are hard to tame.

I'd have to disagree here. Caveat: I don't have PCB building or pressure sensor experience but I did play with low S/N ratio signal aquisition for some physics experiments.

My idea is as follows:

Get pressure sensor output and scale it with op amp to be within 0-5v. This will undoubtly pick up all sort of noise. However the trick is to put lowpass filter right before ADC input. Careful construction is important and the cut off frequency should be as low as possible to improve noise rejection property of the filter. The only noise left will be in the low freq part of the spectrum or coming between the output of the filter and ADC input. If I build the filter directly on the input of ADC chip there is little place for the noise to appear. I presume ref voltage for ADC is clean enough to get 16 bit resolution.

What do you think?

Regards,

Val.

Share this post


Link to post
Share on other sites

That has been my method as well. But, I usually do part of the filtering digitally too.

I have no doubt that your circuit will work with a 16-bit A/D. Heck, get it to settle out with 13 or 14 effective bits and you will probably have more altimeter resolution than you need if you select one of the lower pressure MPX sensors.

I am using the PIC's internal 10 bit A/D on my telemetry project and get better than two foot resolution (I amplify and scale the signals in the analog domain). The pressure signal is noise free, but my digital auto-zero circuit has a little bit of drift. It is a problem that I have yet to get back to.

Edit: I forgot to consider that you fly at very high relative altitudes. You will not have the luxury of using a low pressure sensor like I used. Once you determine the required dynamic range and resolution, you can pick your sensor, analog conditioning, and A/D needs that will support the app.

My main point is to convince you that the design, including the board layout, is very critical in order to effectively eek out 14-16 bits from the A/D. With home brew assembly it can be done with care and a bit of luck (an off-the-shelf A/D board will help a lot).

Edited by Mr.RC-Cam

Share this post


Link to post
Share on other sites
My main point is to convince you that the design, including the board layout, is very critical in order to effectively eek out 14-16 bits from the A/D

Good advice, 12-14 bits should be enough for what I am doing. Thanks.

Val.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×