Cumulus Support

dive76 wrote:I use 40m of CAT5 cable and serial to usb converter. No double tips at all!

I was curious the other day what length this would work on CAT5 / 6

Folks,

1 - Should I put a capacitor and resistor on the RG-11 connected to Cumulus using the RS-232?

2 - Are we dealing with debounce here being an issue with initial methodology of connectivity?

I am wondering though why I do not see this (as far as I know) connecting the RG-11 to the 1-wire counter?

This morning decided to install the 2nd Davis Vantage Vue next to the two RG-11's, connect up a second console / second instance of Cumulus. Will post a picture today or tomorrow.

Number 1 has cured the "on" "on" in my case and several others have posted here with success.

I am also curious as to why other counters have not logged this. Could it be it's so quick, 50ms I believe, that the 1-wire counter can't register that quick?

duke wrote:I am also curious as to why other counters have not logged this. Could it be it's so quick, 50ms I believe, that the 1-wire counter can't register that quick?

Perhaps it's just a result of using the serial port control lines for something they weren't intended for? And devices which were actually designed for this sort of thing handle it properly?

Currently utilize one dual counter for two water meters and lightning sensor. The Dallas tipping bucket I am using for testing hasn't been used in a couple of years. I did configure/set up the Davis Vantage Vue such that it is adjacent to the RG-11's.

The specs on the DLJ water meter (two in place) are:

Contact Rating: 4 Watt
Max Voltage: 24 VDC
Max Current: 0.01 A
Max Contact Resistance: 0.2 ohm
Capacitance: 0.2 pF

Lighting is a per hour count that sometimes goes into the tens of thousands.

Attached is the Maxim specifications sheet for the DS2423. There is some mention of an internal debounce on some owfs development forum.

http://datasheets.maxim-ic.com/en/ds/DS2423.pdf

So I will add the resistor / capacitor in place (to test) on the RG-11 connected to Cumulus. I've seen various values to utilize and am a bit confused. Is there a consensus on the values to utilize?

Is there a consensus on the values to utilize?

Around 800ohms seems to work well.

Thank-you Duke.

Added Davis Vantage Vue #2 to mix. This one will probably go to Florida replacing the Fine Offset in the next few months or so. The two RG-11's and Dallas Tipping bucket are perpendicular and a bit out from the Davis Vantage rain tipping bucket view with the Vantage Vue solar panel facing South. The post itself (flat pieces) face North, South, East and West (very close).

We had a substantial amount of rain over night and into the morning. That said I am seeing about 50% more rain measured on the RG-11 when comparing it to the Dallas Tipping bucket and Davis Vantage Vue.

I will give a try to the resistor installation to see how it works.

Hi,

Note that the "functional" component in the "anti-contact-bounce" circuit is a capacitor of perhaps 10uF. The series resistor with a value of something under 1k ohms is only to (possibly) prevent minor long-term damage to the relay contacts.

However, IMHO it is impossible to "recommend" any optimum circuit because there are considerable variations in COM port "RS232" hardware. "Real" COM ports typically work with + and minus 12 volt levels from a fairly high source resistance whilst a USB-serial converter may deliver 5 volts or less from a much lower impedance. The original RS232 specification mandated + and minus 3 volt threshold (detection) levels, but most/all interfaces now don't require any negative voltage at all.

Cheers, Alan.

Alan,

Thank-you.

The Cumulus connected RG-11 is via an indirect Quatech Serial to IP box.

The server though is sitting about 4 feet from the Quatech box and I could connect the serial port directly if need be. I also utilize Digi Serial to USB devices (have spare 16 and 8 port boxes), generic 4 port serial to USB and numerous serial to USB devices.

The Quatech box is a QSE-100D with the following specifications:

Serial:
RS-232 (EIA-232), asynchronous DTE configured with full modem control and hardware flow control; Configurable as RS-232 or RS-422 or RS-485 (RS-232/422/485)

Port Power:
SSE-100D-5V and DSE-100D-5V provide +5V at Pin 9 on the DB-9 serial port connector
Receiver Inputs:
RS-232:
Input Voltage Rating: -15V to +15V
Receiver Skew: 120 ns (typical), 250 ns (maximum)

Transmitter Outputs:
RS-232:
High Level Output: +5V (minimum), +5.4V (typical)
Low Level Output: -5V (minimum), -5.4V (typical)
Transmitter Skew: 50 ns (typical), 200 ns (maximum)

Receiver inputs (RS-232)
Input Voltage Rating: -15V to +15V
Receiver Skew: 120 ns (typ), 250 ns (max)

UARTs:
SSE: 16C550 with 16-byte FIFO
DSE/QSE/ESE: 16C750 with 64-byte FIFO

Surge Suppression: (I have had one lightning strike around here in about 10 years and it damaged sprinkler controller but not the Digi 8 port serial to USB device).
Surge suppressor applied to each line that is capable of sustaining up to 40A peak, 8 x 20µs transient surges, a clamping voltage of 30V and a peak energy dissipation of 0.1 Joules. (NOTE: The "SS" option limits data rate to 115.2 kbps due to capacitive loading). Standard on 100D-5V models, optional on all other models.

So the capacitor goes across the two RS-232 lines and the resister is in series with which of the two lines?

Drawing attached. Apologies for making a mountain out of a molehill.

I haven't really followed this thread, but...

If you have +12V available at the RG11, could you not use that as your reference voltage (connect +12V to 'Common') rather than looping back DTR, which will normally only come high once Cumulus has taken control of the port and potentially could cause a miscount at start-up?

Edit: I guess that means that the grounds must be reasonably 'close' to each other.

Thank-you Duke. I saw something earlier about using an electrolytic capacitor but wasn't totally sure.

So for Cumulus RG-11 "tipping bucket" mode I only need to use DTR and RTS?

If I am using the RG-11 in "rain" mode then I use DTR & RTS / DSR & CTS lines eh?

pete_c wrote: Configurable as RS-232 or RS-422 or RS-485 (RS-232/422/485)

Hi pete,

That's one reason why I'm cautious generalising about "RS232" interfaces. IMHO your interface diagram is theoretically preferable (but with the capacitor polarity as shown by duke).

Mark is correct that using an "external" supply rail would remove one "unknown" from the system, and in practice even a 5 volt rail should be sufficient. The USB-serial adapter that I use (successfully) outputs only +5 volts to assert DTR.

Where the circuit is not compliant with (the original) RS232 specification is that there is no pull-down resistor to a negative rail, but most/all present interface circuits appear to accept "earth" potential in place of the negative threshold. However, perhaps there should be a resistor of 10k - 22k from DSR and/or CTS to earth, to better define the non-asserted (relay contacts open) signal level.

With the latest version of Cumulus (1042) "Tips" and/or "Raining" can be signalled on DSR and CTS in either combination.

Cheers, Alan.