If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Also - it seems a missed opportunity that the mains powered modules (QLH) don't supply the IDRANet power... Is there any way of having dual power supplies onto IDRAnet - such that one is redundant ? Maybe I could do this via a couple of diodes and a slightly raised PSU voltage ?
Presume you mean QLD? We have a dedicated IPD module whose main function is to monitor and handle power delivery to the network and to condition/charge a lead acid battery backup. An additional 5A SMPS unit is required to deliver the actual power to the IPD. I'm not sure what the motive is for getting power via other mains connnected modules, but in any case it will have practical problems... It is a challenge to squeeze in the electronics and get 1KW out of the QLD in a 6M DIN box whilst managing to dissipate enough heat. Adding a power supply will significantly increase size and cost per unit and you would not get other functional aspects of a module like the IPD.
Yes you can have multiple supplies switched by diodes as you suggest.
Karam - can I perhaps reflash the LTH to allow for pulse counting? How did the anemometer example work? Do I have to put in my next order for more parts? ;-)
John
A beta update to the LTH firmware is now available to introduce pulse counting on the digital inputs. Anyone else interested can e-mail me directly ([email protected]) to request this and instructions. Updates also available for PLH, LTS and PLT modules.
OK, we'll be wanting to start wiring in a month or two, so 'thought I'd prod this thread - maybe someone's bought a few 'meters & tried them ... before I try to order some ! ?
So I plan to start by pulse counting the LED flashes from the nice new meter that was kindly installed recently, but I have a great interest in whether anyone is using any of the 1-way DIN rail counters for individual circuit monitoring.
And finally, before I shop around, how many people would be up for a bulk buy and in what quantity?
I trialled a couple of different 1-DIN units and they all worked well with Cortex pulse counting. The nicest ones seem to be those that can pulse at 1W/h so 1,000 times for a KWh as they give virtually real-time power.
Definitely up for a group buy I'd need 8 maybe 10 to be safe. The other issue is how to fit them in the consumer unit? I have seen a double banked CU so normal RCD / MCB occupy one row and the counters the second row?
Then of course I'd want to look at gas and water supply ;o)
good ... not seen any that give 1000 per kWh ... so good again ...
yep, was thinking ten, too ...
need to think current capacity, of course, too, but if they're all 45A+ (like the ones early in this discussion) then I guess no problem ...
>resurrected this discussion ...
yes, that was in association with the Current Cost approach, of course, for which a new model looked to be about (http://dalelane.co.uk/blog/?p=389) to offer overall plus (sometime in the future) nine individual monitoring capability & multiple displays ... and possible linking to Cortex via RS232 & xAP ...
actually, that's a question, presumably the new DFP unit could be used to display derived power usage information ?
>gas & water ...
yep, me too !
Last edited by chris_j_hunter; 16 January 2009, 11:35 AM.
Reason: typos
I use the Din units along with a couple of the the bigger full rated standalone ones. I also use Current Cost via xAP but haven't got that latter bit integrated into Cortex's 'power' world yet. Very keen to look at the alertme smart plug type Zigbee modules too - due this month.
One of the issues - particularly if you had pulse outputs at every WH is that the counter that the IDRANet devices implement is only 8 bit and so has to be read and cleared by Cortex within circa 200 counts - with several metering circuits and high energy levels this would generate a lot of IDRAanet traffic. It also becomes inaccurate every time Cortex is stopped eg for updates.
Having a 16 bit or ideally 24+ bit counter would be really nice as then the hardware count could effectively mirror the displayed count on the meter. But I know memory space is very tight in the modules.
I am getting slightly nervous / annoyed at Alertme as I have now sent two emails requesting to be put on the waiting list and get the intial discount as advertised. In addition I asked for some technical information concerning the units and now both emails have gone unanswered.
With Alertme and Current Cost are we confident that both current and voltage is being measured to give true power? Or is it a case of derived power whereby the current is measured through a current clamp and the voltage is "assumed" to be 230vac? In the case of an inductive load this will be incorrect.
In the case of the ODI doesn't it bunch the pulse counts into a IdraNet packet and then send packet only once every 60 seconds or something?
interesting - does this mean we should see Idratek & Alertme as essentially complementary systems ?
it'll be interesting to see how much they can be integrated together ... the mobile nature of Alertme could be handy in some situations - eg: to protect / control temporary works (protect with children as well as the uninvited) ...
bit nervous about battery costs with Alertme, though, even with the batteries available at about half the normal cost from the likes of Directfoto ...
Last edited by chris_j_hunter; 16 January 2009, 05:13 PM.
I believe Alertme have released SDK so hopefully someone like Kevin can develop a connector for xAP. Once it is xAP enabled it would be pretty easy to get into Cortex.
BTW, I've updated my website to include Kevin's worked example of getting a XAP object into Cortex. Currently part 1 is available which discusses the pre-reqs. I've written up part 2 (discussing the actual software configuration of getting Cortex and Xap connected) and just need to proof read and check formatting before publishing.
One of the issues - particularly if you had pulse outputs at every WH is that the counter that the IDRANet devices implement is only 8 bit and so has to be read and cleared by Cortex within circa 200 counts - with several metering circuits and high energy levels this would generate a lot of IDRAanet traffic. It also becomes inaccurate every time Cortex is stopped eg for updates.
Having a 16 bit or ideally 24+ bit counter would be really nice as then the hardware count could effectively mirror the displayed count on the meter. But I know memory space is very tight in the modules.
If you are drawing 3.6KW through each meter (16A) then I think that's a pulse a second or a packet dispatch every 3.3 minutes or so. 20 meters running at 3.6KW each (320 amps in total) and its a packet every 10 seconds. Not so elegant but would not exceed IDRANet bandwidth by any stretch - though if you are drawing that sort of power you may have a large property and maybe numerous IDRATEK modules as well . The metering function was intended for day to day indication rather than a full cumulative record - ie. so that people could see how they were using electricity in day to day tasks and get perhaps warnings and so on - rather than give them a highly accurate figure of electricity consumption over a month say.
However if either of these issues prove to be a real problem in practice we can no doubt take necessary action relatively easily. For example dedicating an ODI to higher resolution reading and perhaps even memorising data in it over a period of time for greater cumulative accuracy. A matter of development priorities really.
With Alertme and Current Cost are we confident that both current and voltage is being measured to give true power? Or is it a case of derived power whereby the current is measured through a current clamp and the voltage is "assumed" to be 230vac? In the case of an inductive load this will be incorrect.
Paul
Taking a guess (and hopefully someone will be less lazy than me) I'd say current cost measures current only since it uses an inductive clamp, whereas Alertme should be able to measure proper power since the voltage signal will be available at the socket adapter - and low cost dedicated ICs to do all the measurements when you have access to the 'juice' are readily available. Does anyone know if the Alertme socket is the what I once saw reveiwed as the 'Plogg'? Seem to remember the latter having Zigby. Ok I'm being lazy again ...
If Alertme ever bother to reply to my emails then I will be in a better position to comment. If power measuring is commonly available in IC then we don't we have an Idratek module?
Comment