PDA

View Full Version : which stamp might be best for a project



Radioman
11-16-2006, 11:24 AM
I need some advice as to which stamp might be best for a project and I have no experience with stamps!

I need to monitor a voltage to meet the following parameters:
A digital input going LOW is the trigger. The voltage must be above a setpoint 50ms after the trigger goes low and must stay above the trigger while the digital input is low.
A second digital input going low will disable the monitor as there is a fault.
A third digital input going low will disable the monitor as there is a second fault.
I would like to output to a LCD or LED readout the voltage a read.
A fault will have to latch an output pin and turn an LED on.

Will the Basic I fit this bill? I got a bunch of um to make so I could get cost prohibitive.

Post Edited By Moderator (Chris Savage (Parallax)) : 11/16/2006 3:55:12 AM GMT

Mike Green
11-16-2006, 11:41 AM
The logic is very simple. The major questions are: 1) what's the voltage range? 2) how accurate does the 50ms have to be? 3) how often do you need to sample the voltage once you reach the setpoint (how quickly would it change)? 4) what kind of LCD or LED readout?

If you really do have to monitor the voltage pretty much all the time, you probably will need an SX or Propeller processor since the Stamps don't do multiple tasks at the same time. Both the SX and the Propeller can do moderate speed analog to digital conversion with only a few passive components.

If things don't change quickly, you could use a Stamp, probably with a simple cheap external analog to digital converter like the ADC0831. Any Stamp could work with a serial input LCD display. The Stamp BS2p series have built-in statements to handle a parallel LCD display (which is very slightly cheaper than a serial one). If the voltage you're monitoring doesn't change quickly, you can monitor it directly with a Stamp (see <http://www.emesystems.com/BS2rct.htm#B_voltage>), but an ADC is more accurate.

The choice depends a lot on the things you haven't mentioned yet about speed needs (which affects the need to multi-task).

Radioman
11-16-2006, 12:03 PM
Monitor needs to:

Monitor 2 analog voltages and 2 digital lines. It must log

faults and report some faults to an alarm reporting system..

The first analog voltage input, AI1, is the power supply to

the Power Amp. This line needs to be monitored for presence and

stability. If the voltage drops below a set point (X1) then

this is a fault and all other inputs are ignored. This analog input can be replaced with a digital one where there are external components to allow the input to toggle around a set point. It would be far better to be able to change the setpoint in software.

Transitions of this line above and below set point X1 should be

logged to a PS FIFO log with time & date.

The first digital input, DI1, is the emergency power input.

If this line is active, the repeater is running on backup

power and the power output of the PA's is drastically reduced.

If this line is active, all other inputs are ignored.

Transitions of this line should be logged to a EP FIFO log

with time & date.

The second digital input, DI2, is the repeater PTT line. When

this line is active, the transmitter has been keyed.

The second analog voltage input, AI2, is the power output

sensor. This voltage is proportional to the actual power

output of the PA. This is the main thing we need to be

monitoring. If the power supply voltage has been above the

set point, X1, for at least 1 second AND the emergency power

first digital input DI1, is NOT active AND the PTT second

digital input DI2 has been active for XXms (X2) then the second

analog voltage AI2, must be above set point X3. If the voltage

is not above set point X3, then this is a fault. Also, during

the time that the second digital input DI2 is active, if the

PA voltage AI2 falls below the set point X3, this is a fault

UNLESS the power supply voltage has dropped below X1 OR the

emergency power input has become active. Faults meeting these

conditions should be logged to a LP FIFO log with time & date

AND should initiate a failure message to the alarm reporting

system.



Now, this is the original specification. I think some of this can be dispensed with but the basic idea is the same. In answer to the previous post:

The voltage range sampled is 0 to 5 volts

The timing is not really critical since the circuits producing the delays are hardware TTL. I have not measured the delay directly yet but since it is a program variable I can work around it.

The voltage must not drop below the setpoint for more than 300 ms and I think that is well within the range of a simple test in the main loop. However, I have not checked the math.

A two digit readout would be just fine but it is the actual voltage I am looking for but the wattage of the Power Amp which is proportional to the voltage. ie: a translation like 1.2 volts might = 20 watts

Mike Green
11-16-2006, 12:50 PM
There's no problem doing a proportional value translation (like 1.2V = 20 on the display). For that matter, doing a non-linear translation using a lookup table is no problem at all with any Stamp. Any Stamp 2 should be able to monitor a slowly changing voltage 3 to 4 times a second using the technique in the link I mentioned. I'm not sure whether the BS1 can do that since the statement involved (RCTIME) is different on the BS1 (POT).

From a cost standpoint, particularly if you're going to make your own PC boards, it's hard to beat the SX or Propeller. For your application, the Propeller protoboard is expected to cost about $15-20 and include a regulator, crystal, and EEPROM for the program. It's hard to beat that price. You'd need a few resistors and capacitors for the A-D converters.

Either the SX or the Propeller could drive 7 segment LEDs directly with just some current limiting resistors needed.
Don't forget that you can display a few letters using the 7 segments like c, d, E, o, A, F, H, J, Y, and maybe some others.