Thank you for reading this.
While looking for something to do recently, I came across a project that I was working on four years ago. It's the basis of a PIC controller for a home-built satellite dish rotator.
It's taken me the better part of a week to figure out what it's all about and I now remember one of the problems that slowed the pace of the project. I'm wondering if it's worth the effort of continuing.
The PIC was counting pulses reasonably well from the elevation jack but was not always accurate because of noise pulses from the motor. I expect the larger azimuth rotator motor will produce even more hash. An accurate pulse count is important so as not to accumulate a large error towards the end of a satellite pass. Constant stopping and starting would probably add extra pulses too.
Do home TV dish pointers use shielded cable to take the position pulses back to the controller? It's not something that I can test myself because my satellite gear is in storage more that 1100km from where I live.
I'm having second thoughts about the pulse counting scheme and I'm beginning to think that the tried and proven ten-turn pot method might be more fool proof.
Has anyone else had any success, or otherwise, with pulse counting? It's an interesting project but it might be impractical.
Regards, Phil.
----- Original Message ----- From: "Phil" phillor@telstra.com To: amsat-bb@amsat.org Sent: Friday, April 25, 2008 1:46 AM Subject: [amsat-bb] Dish position pulse counter
Thank you for reading this.
I'm having second thoughts about the pulse counting scheme and I'm beginning to think that the tried and proven ten-turn pot method might be more fool proof. Regards,
Phil.
Hi Phil
The ten-turn pot method is also not immune from inaccuracy when the voltage across the pot is close to 0 volt i.e. when the rotator is turned near to 0 degrees elevation. In this condition the noise voltage picked up by the cable running from the antenna to the control box can add with the low voltage available across the potentiometer. Since you are able to write a program for a PIC a better method to get in general a high pointing accuracy requested for a dish and expecially for a big dish is to use different type of position sensors like the following:
Sequential encoders Absolute encoders Selsyn or Resolvers
With two inexpensive surplus selsyn for the azimuth and two selsyn for the elevation on my dish I actually get a very accurate indication of position on mechanical pointers but adding the Analog Device chip AD2S80 I plan to write a program to get both precise digital indication and automatic tracking with high accuracy
Look on google for the AD2S80 data sheet and for many technical applications of it with syncro-resolvers.
Have fun
73" de
i8CVS Domenico
If pulse counting is used and since presumably the duration and expected repetition of the pulse is known / can be calculated is it not possible to discriminate between noise and a valid pulse?
-----Original Message----- From: amsat-bb-bounces@amsat.org [mailto:amsat-bb-bounces@amsat.org] On Behalf Of i8cvs Sent: 25 April 2008 11:11 To: Phil; AMSAT-BB Subject: [amsat-bb] Re: Dish position pulse counter
----- Original Message ----- From: "Phil" phillor@telstra.com To: amsat-bb@amsat.org Sent: Friday, April 25, 2008 1:46 AM Subject: [amsat-bb] Dish position pulse counter
Thank you for reading this.
I'm having second thoughts about the pulse counting scheme and I'm beginning to think that the tried and proven ten-turn pot method might be more fool proof. Regards,
Phil.
Hi Phil
The ten-turn pot method is also not immune from inaccuracy when the voltage across the pot is close to 0 volt i.e. when the rotator is turned near to 0 degrees elevation. In this condition the noise voltage picked up by the cable running from the antenna to the control box can add with the low voltage available across the potentiometer. Since you are able to write a program for a PIC a better method to get in general a high pointing accuracy requested for a dish and expecially for a big dish is to use different type of position sensors like the following:
Sequential encoders Absolute encoders Selsyn or Resolvers
With two inexpensive surplus selsyn for the azimuth and two selsyn for the elevation on my dish I actually get a very accurate indication of position on mechanical pointers but adding the Analog Device chip AD2S80 I plan to write a program to get both precise digital indication and automatic tracking with high accuracy
Look on google for the AD2S80 data sheet and for many technical applications of it with syncro-resolvers.
Have fun
73" de
i8CVS Domenico
_______________________________________________ Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
Phil,
Have you identified the source of the noise? However you regulate down the 5v for the pic, if you place a larger capacitor on the 5v regulated bus (between 5v and gnd) this might help with these fluctuations, essentially acting as a high pass filter.
If the power bus is clean, the noise might be induced along a long cable run to the rotor itself. It might be worthwhile to look at a differential a2d with a spi or i2c interface.
Take a look at the Texas Instruments ADS1110. It has an internal reference voltage as well to help filter out these ratiometric effects.
Another option might be to switch over to a rs485 or rs422 link between the rotor and your shack where the controller presumably is.
-Phelps KI6GST
On Fri, Apr 25, 2008 at 6:29 AM, David Barber < david.barber@dbelectronics.co.uk> wrote:
If pulse counting is used and since presumably the duration and expected repetition of the pulse is known / can be calculated is it not possible to discriminate between noise and a valid pulse?
-----Original Message----- From: amsat-bb-bounces@amsat.org [mailto:amsat-bb-bounces@amsat.org] On Behalf Of i8cvs Sent: 25 April 2008 11:11 To: Phil; AMSAT-BB Subject: [amsat-bb] Re: Dish position pulse counter
----- Original Message ----- From: "Phil" phillor@telstra.com To: amsat-bb@amsat.org Sent: Friday, April 25, 2008 1:46 AM Subject: [amsat-bb] Dish position pulse counter
Thank you for reading this.
I'm having second thoughts about the pulse counting scheme and I'm beginning to think that the tried and proven ten-turn pot method might be more fool proof. Regards,
Phil.
Hi Phil
The ten-turn pot method is also not immune from inaccuracy when the voltage across the pot is close to 0 volt i.e. when the rotator is turned near to 0 degrees elevation. In this condition the noise voltage picked up by the cable running from the antenna to the control box can add with the low voltage available across the potentiometer. Since you are able to write a program for a PIC a better method to get in general a high pointing accuracy requested for a dish and expecially for a big dish is to use different type of position sensors like the following:
Sequential encoders Absolute encoders Selsyn or Resolvers
With two inexpensive surplus selsyn for the azimuth and two selsyn for the elevation on my dish I actually get a very accurate indication of position on mechanical pointers but adding the Analog Device chip AD2S80 I plan to write a program to get both precise digital indication and automatic tracking with high accuracy
Look on google for the AD2S80 data sheet and for many technical applications of it with syncro-resolvers.
Have fun
73" de
i8CVS Domenico
Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
Sent via AMSAT-BB@amsat.org. Opinions expressed are those of the author. Not an AMSAT-NA member? Join now to support the amateur satellite program! Subscription settings: http://amsat.org/mailman/listinfo/amsat-bb
participants (4)
-
David Barber -
i8cvs -
Phelps Williams -
Phil