Stamp pulser
Archiver
Posts: 46,084
For those that are interested in the programming aspects of a Stamp
pulser and the Stamp enhancements, the following programming/hardware
suggestions are provided.
Use a 10K NTC thermistor epoxied to a TO-220 heatsink and mounted on the
FET. Use a 10K NTC thermistor exposed to ambient air temps.
Connect them in the unpreferred configuration of the RCTIME example with
a high voltage rated 0.1uF cap to ground, the thermistor to VDD. This
enables reading the charge time from ground to noise instead of the
discharge time to some noisy spike.
Calibrate the thermistor readings in order that the FET thermistor
returns a value slightly lower than the ambient air thermistor with the
RCTIME command. Divide the result by 10.
Read the thermistors
If less than 10 difference in readings then increment power
If more than 10 difference in reading then decrease power
Increasing/Decreasing power can be done by one of two means
1. Holding the FET pulse width to 50uSecs (12V) and varying PRF by
changing a dummy PULSOUT to an unused I/O pin
2. Starting with a minimum PULSOUT of 1 to the FET and increasing until
the difference temperature is reached.
In all cases do not allow the pulsing loop (FOR-PULSOUT-NEXT) to exceed
more than 5 to 10 seconds in duration before you check thermistor temps
(20,000 to 40,000 loops). The FET can heat up quickly and smoldering
polyolefin covering on the inductors smells terrible. [noparse]:)[/noparse]
The second method yields a pulser that will automatically set itself to
pulse anything between 6V and 24V.
It works as the FET will begin to heat up when saturation of the
inductor is near or has been achieved. This affords MAX pulse voltage
out as it runs at the edge of the inductor performance envelope.
Expect to see pulse voltages in the neighborhood of 30V to 80V (or more
on a really bad batt). They last 0.1uSec with some decaying
oscillations.
And to anticipate the question that is ALWAYS asked,
NO it will not harm connected equipment. Whether it is an inverter or
automotive computer. All equipment has transient voltage protection.
Inverters due to nearby lightening strikes and automotive computers due
to the hash that can be generated by a generator or alternator with a
bad bearing.
The 100nSec pulse does not have enough energy in it to harm anything.
Most underrated components will survive due to the reaction time of the
underrated component.
Well I think I covered Pulsing Stamp basics.
Happy pulsing,
Don
pulser and the Stamp enhancements, the following programming/hardware
suggestions are provided.
Use a 10K NTC thermistor epoxied to a TO-220 heatsink and mounted on the
FET. Use a 10K NTC thermistor exposed to ambient air temps.
Connect them in the unpreferred configuration of the RCTIME example with
a high voltage rated 0.1uF cap to ground, the thermistor to VDD. This
enables reading the charge time from ground to noise instead of the
discharge time to some noisy spike.
Calibrate the thermistor readings in order that the FET thermistor
returns a value slightly lower than the ambient air thermistor with the
RCTIME command. Divide the result by 10.
Read the thermistors
If less than 10 difference in readings then increment power
If more than 10 difference in reading then decrease power
Increasing/Decreasing power can be done by one of two means
1. Holding the FET pulse width to 50uSecs (12V) and varying PRF by
changing a dummy PULSOUT to an unused I/O pin
2. Starting with a minimum PULSOUT of 1 to the FET and increasing until
the difference temperature is reached.
In all cases do not allow the pulsing loop (FOR-PULSOUT-NEXT) to exceed
more than 5 to 10 seconds in duration before you check thermistor temps
(20,000 to 40,000 loops). The FET can heat up quickly and smoldering
polyolefin covering on the inductors smells terrible. [noparse]:)[/noparse]
The second method yields a pulser that will automatically set itself to
pulse anything between 6V and 24V.
It works as the FET will begin to heat up when saturation of the
inductor is near or has been achieved. This affords MAX pulse voltage
out as it runs at the edge of the inductor performance envelope.
Expect to see pulse voltages in the neighborhood of 30V to 80V (or more
on a really bad batt). They last 0.1uSec with some decaying
oscillations.
And to anticipate the question that is ALWAYS asked,
NO it will not harm connected equipment. Whether it is an inverter or
automotive computer. All equipment has transient voltage protection.
Inverters due to nearby lightening strikes and automotive computers due
to the hash that can be generated by a generator or alternator with a
bad bearing.
The 100nSec pulse does not have enough energy in it to harm anything.
Most underrated components will survive due to the reaction time of the
underrated component.
Well I think I covered Pulsing Stamp basics.
Happy pulsing,
Don