GEM pulse counters....loosing counts.....a possible bug

[mminehan]

So...I have been going insane trying to work out why my pulse counters keep loosing counts for my water and gas meters (both contact closure reed switches).

I have spent many hours over several months trying to sort this out, even building a pulse divider circuit to try and solve the problem, thinking the pulse duration was too short. I have tried channels 3 & 4 (simple contact closure) and channels 1 & 2 with a 12v pulse. In both situations both the water and gas meter being to loose counts over time, even with the pulse divider that bench tested perfectly.

Sometimes the water meter would not count at all. Time and time again I checked the wiring and conductivity. And every time my multi meter would detected the contact closure without problem. I thought I must be going nuts. What is going on?

Today I stumbled upon what I think is the problem: if one channel is closed (in the case or 3 & 4) or high (in the case of 1 & 2) the other channel can not be read. What is happening is that if either meter stops when the contact is closed the other meters pulses are not counted.

To test this theory I wired just the water meter to channel 4 and it counts perfectly. But as soon use a jumper to close channel 3 the water meter on channel 4 stops counting.

I then did the same for channels 1 & 2. If I apply 12v to either channel 1 or 2 the other channel can not be counted.

Interestingly this only happens in pairs (1&2 and 3&4). Applying 12v to channel 1 or 2 does not stop 3 or 4 form working.

Here are my system details:
COM firmware ver: 4.0
ENG firmware ver: 1.49
RealTime Status: ON
Packet Send Interval: 15
Pri. Packet Format: 5
Sec. Packet Format: 0
Sys Status: OK

Can some else please check this on their system? Try jumping closed channel 3 or 4 and see if you can detect pulses on the other channel (4 or 3 respectively).

In the mean time I am only going to connect just the water meter and see what happens over the next few weeks......just to make sure I'm not barking up the wrong tree.....or still insane...lol.

Any help much appreciated.

Thanks, Marty.

[mminehan]

So....just an update. With only one meter connected (the water meter) the pulses are being counted perfectly.

I'm more certain now that if both channels on 3 & 4 (or 1 & 2) are closed at exactly the same time that only one of channel's pulse is counted. The other is ignored leading to the inaccuracies.

I would really appreciate if someone could verify this for me.

Marty

[ben]

Paul is taking a look into this. He says to work around it for now, put one on channel 1 and one on channel 4.

[mminehan]

Hi Ben,
Has there been any progress with this? Has Paul noticed the same issue?
Marty

[ben]

Hi Ben,
Has there been any progress with this? Has Paul noticed the same issue?
Marty

No progress as of yet, but Paul was able to reproduce it.

[mminehan]

Ok...that's cool. I just wanted to make sure I wasn't going mad.

I hope it is just a firmware thing and not a hardware limitation.

Marty

[mminehan]

Has there been any progress on this?
I would prefer not to have use an external power source (i.e. ch 1 or 2) to get one of the meters to work.

Can you use the +5VDC power out on the board for this purpose?

Marty

[Teken]

Hello Marty,

Brultech is testing out a new COM 4.14 firmware now to see if it resolves the pulse issue. I have this firmware loaded now on my system.

I will try to find some time to test it out and report back pro's / con's.

[mminehan]

Teken,

That's great news. Thanks for working on this and the update. Look forward to hearing about your testing.

Regards, Marty.

[ben]

Teken,

That's great news. Thanks for working on this and the update. Look forward to hearing about your testing.

Regards, Marty.

I've emailed it over to you as well if interested.