RF video reciever - digital tuning

[Jumper]

Hi,

A budget solution:

Use the Hardware PWM in the PIC together with a filter (2 caps and 2 resistors) to get an adjustable DC level by changing the duty cycle of teh HPWM. Feed this in to an OP if you need more current. This is excactly the same as having a 255 step DA converting chip by only adding a few cents of components. If you dont have a HPWM, it is possible to make a SW one but pick a PIC that has.

Buy a LM1881 (sync separator chip) for a dollar or two, connect the video source from the receiver to it. Then connect the V-sync signal from this chip to your PIC.

Let the software check the V-sync signal period to make sure it is correct, if it is not correct, adjust the duty cycle and voila ,hopefully :-)


Then you have a kind of PLL, since you are locking the HPWM dutycycle to the v-sync signal from this chip. The LM1881 has been around for a long time so it is nothing special.


Total cost less then 10 USD with PIC and all.

/me

[Ron Marcus]

Hi,

A budget solution:

Use the Hardware PWM in the PIC together with a filter (2 caps and 2 resistors) to get an adjustable DC level by changing the duty cycle of teh HPWM. Feed this in to an OP if you need more current. This is excactly the same as having a 255 step DA converting chip by only adding a few cents of components. If you dont have a HPWM, it is possible to make a SW one but pick a PIC that has.

Buy a LM1881 (sync separator chip) for a dollar or two, connect the video source from the receiver to it. Then connect the V-sync signal from this chip to your PIC.

Let the software check the V-sync signal period to make sure it is correct, if it is not correct, adjust the duty cycle and voila ,hopefully :-)


Then you have a kind of PLL, since you are locking the HPWM dutycycle to the v-sync signal from this chip. The LM1881 has been around for a long time so it is nothing special.


Total cost less then 10 USD with PIC and all.

/me

I like the idea of this, but how does the PIC know which way to tune, if it loses the sync? I bet that there is an AGC point on the circuit board, which will give a measurable voltage change with signal level. This might give a better tune indication than the sync pulse. If you get a D to A chip which is a resistive network, rather than a charge pump, your noise level will be much easier to minimize. You could also use the eight lines of an I/O register to create an R2R network to accomplish the same thing with lower noise.

Ron

[Jumper]

When the PIC senses that the signal has been lost (or getting bad) it could try to tune up a certain number of steps, if that didn't work it tunes down instead. The signal must be somewhere, or it just sweps the entire band for a signal.

Probably will the FQ drift mostly in one direction and then we start tuning that way.

Feeding the AGC point into an AD converter in the PIC would be neat but it requires som reverse engineering to find it.

A proper filtered HPWM should have a nice DC level out but as you say, a resistor network from the I/O will do just as fine.

[RYTECH]

I've got my D/A chips from maxim.. i'll report back if the digital tuning solution using D/A works out...

[RYTECH]

tuner works using max518's..

the 8 bit DAC is sufficient for tuning in the various frequencies.