Sandaysoft

The test program shows a triple 'on'. This confirms that it's not an issue with the serial component which Cumulus uses, and these state changes are really occurring, so Alan's mods are necessary.

_________________
Steve
Sanday Weather
----------------------------------------------------------------------------------------------------------------------------------
Like Cumulus and want to support it? Please donate!

just thought it was best to try this to confirm one way or another as to what was happening.

going to stick the cap back on and rerun the test







i think we might just about cracked it

_________________
http://www.wunderground.com/weatherstation/WXDailyHistory.asp?ID=IARGYLLA4

Hi
I'm happy - not exactly the most complicated circuit



Regards
Alan

Steve would it be worth doing the same test on the old dell with the direct rs232 port or would you rather just have this running on the usb to rs232 convertor?

_________________
http://www.wunderground.com/weatherstation/WXDailyHistory.asp?ID=IARGYLLA4

Hi Bruce

As RS232 ports are becoming quite rare IMHO stick with the adapter. Having said that if my mod is curing relay bounce then there seems to be no reason that an rs232 would react differently.
Regards
Alan

It would be useful to know whether the solution works with a real serial port, but being practical I think it would be reasonable to say that if it doesn't, then a USB adapter should be used.

_________________
Steve
Sanday Weather
----------------------------------------------------------------------------------------------------------------------------------
Like Cumulus and want to support it? Please donate!

Hi Alan

the reason for trying the direct port is more to do with i want to run the old dell to monitor things when im on holiday.
have given it a quick try and looks ok. some users might be running older pc's which have the rs232 port so really should give them the option to compare results on both the direct rs232 and the usb convertor.

Bruce

_________________
http://www.wunderground.com/weatherstation/WXDailyHistory.asp?ID=IARGYLLA4

1/ have we concluded that the problem that has apparently been solved was relay bounce?

2/ is it worth trying other value resistors as per Ole's recommendations or stick with what is now working?

3/ if we stay with this value could this cause any unseen problems later? I think probably only Ole will be able to answer that.




Sent from my Windows phone.

_________________
Duke

Nightingale Weather PHP
Eastleigh, Hampshire, UK

Hi
Relays are expected to work for about 1,000,000 cycles so with average rain (5mm per day) that would be about 20,000 cycles (an ON and an OFF) so even if the bounce spark degraded the relay life to 20% that's still about 10years of life by which time there may be something else to think about. And we really don't know if it is bounce.
Regards
Alan

Alan,

Would you mind sharing a schematic drawing of your circuit? Would help me to build the same here and try.

Thanks,

Lars

Hi Lars

I said it was complicated
I made a mistake in the drawing - reverse the positions of the resistor / capacitor - so resistor from +ve to DTR



Regards
Alan

Alan, are you in a position to test this with RTS/CTS?

_________________
Duke

Nightingale Weather PHP
Eastleigh, Hampshire, UK

Hi
I have some old female 9 pin leads so I guess I could tack one on when I get a moment.
Alan

Hi
Your wish is my command
RTS/CTS result (same Cap/res config)
Apart from a starting OFF the debug looks OK (Again very heavy hosepipe rain)

Alan

Alan, you're a gentlemen, thank you.

I think you've also now earned a title, "Cumulus test engineer"

Perhaps, we should also now include your new rain description in current conditions "Again very heavy hosepipe rain"

The resistor, just 1/4w?

_________________
Duke

Nightingale Weather PHP
Eastleigh, Hampshire, UK