Does the emonTx Shield need an actual Arduino?

fabiopbx · 2016-12-22T17:27:56.226Z

pb66:

But if you put all the CT’s on the same phase as the AC:AC adapters you could compare them, side by side and confirm they are all working “similarly” rather than “just working” but potentially very different.

very true ON MY WAY! lol

pb66 · 2016-12-22T18:03:33.341Z

5 posts were split to a new topic: Enabling Redis is causing issues in emonCMS

fabiopbx · 2016-12-22T17:37:25.486Z

ermm… so CT2 is negative cause its flipped if im not wrong but has a similar PF if we dont count the negative value…

CT3 i have no idea…

EDIT: ohh wait that has no Amp reading… ish

blob.jpg1765×363 310 KB

EDIT2 - same AC-AC phase, all CTs into same phase

Pasted image1486×352 214 KB

pb66 · 2016-12-22T18:09:27.849Z

fabiopbx:

same AC-AC phase, all CTs into same phase

That looks much better power factor wise, but the other values are very inconsistent so it’s difficult to compare, do you have a steady load you can monitor with the 3 CT’s or can you use emonhub or similar so that all 3 nodes results are simultaneously shown.

fabiopbx · 2016-12-22T18:14:32.647Z

Steady load i dont think i have anything lying around tbh…

those are live feeds from the serial output, they should be very simultaneous accounting for my fast screen /dev/ttyUSBX copy paste skills

but from emonhub i can paste too:

Pasted image1812×646 32.8 KB

pb66 · 2016-12-22T18:24:15.883Z

fabiopbx:

those are live feeds from the serial output, they should be very simultaneous accounting for my fast screen /dev/ttyUSBX copy paste skills

but from emonhub i can paste too:

not to worry, The fact they are all running on 10sec intervals and not synced will result in variation anyway.

fabiopbx · 2016-12-22T18:25:33.429Z

would sync be a good idea?

and how would that be accomplished if so?

pb66 · 2016-12-22T19:41:53.469Z

It would be a good idea but it’s not easily accomplished with what we have, emonhub would need to poll for the info or a “trigger signal” would be needed, either way the sketches would also need changing.

For the moment the nearest you could get is to power cycle (not reboot) the Pi so that the 3 emonTx’s all start at the same time, it will go out of sync over time but to be honest with the load you have varying so much and a discreet sampling sketch they would need to be absolutely in sync to be directly comparable, hence the “better to test on a constant load”.

Dave · 2016-12-22T21:22:46.500Z

I use a 2Kw heater and a long phase cable to wrap through the CT a couple of times, it will only vary slightly then as the voltage creeps up and down.

Paul, the ones I bought are…

eBay

DL-CT08CL5 20A/10mA 2000/1 0~120A Micro Current Transformer | eBay

Linear protection current: 0 ~ 120A. Rated output current: 0 ~ 10mA. Rated Input Current: 0 ~ 20A. Zero sequence / drain / remaining capacity detection and monitoring.

Regards
Dave

pb66 · 2016-12-22T23:18:54.931Z

Dave:

I use a 2Kw heater and a long phase cable to wrap through the CT a couple of times, it will only vary slightly then as the voltage creeps up and down.

That’s the easiest way to test, you can also test at various increments just by adding or removing CT passes.

Looking at those CT’s I’m guessing you will be extending the wires, if it were me I, would factor the diodes in at that point rather than putting them in the jack connector, but I don’t think it’s overly important.

Dave · 2016-12-23T10:22:57.690Z

Yeh I was going to twist decent alarm cable together and solder with heatshrink to extend.
I’m hoping these work well as these are a lot smaller than the standard CTs and with 12 CTs in my fuseboard, its cramped to say the least.

Regards
Dave

Robert.Wall · 2016-12-23T13:27:46.265Z

pb66:

Is it a good idea to put a zener diode in for safety if your not using standard CTs?

I’d prefer Robert to answer that as it is a safety issue, I usually have in the past but not always. putting safety precautions in place is generally speaking always a good idea.

A CT is the dual of a voltage transformer. If you short-circuit a (normal) voltage transformer, you get excessive current and it burns out. A current transformer does the opposite: If you open-circuit a current transformer, you get excessive voltage (as it tries to drive the current it generates into a near-infinite impedance) and it flashes over and destroys the insulation. So, as I’ve written may times in the past, a CT can always be short-circuited in safety, because then it is delivering zero power.

Therefore, there are two things you can do:

Short-circuit the CT before you unplug/disconnect it, or
Fit a pair of voltage suppressors (or one bidirectional one), or a pair of zener diodes in series, to clip the voltage at some safe value.

It’s not likely that a very small CT like the ones we use would kill, but it could easily be very painful, and would certainly destroy any delicate electronics that couldn’t handle the current.

{Sorry I’ve been away for a day - I’m back on-line now.}

Robert.Wall · 2016-12-23T13:33:18.795Z

fabiopbx:

it seems either my soldering is bad (looks good tbh) or the jacks are faulty…

I have seen faulty jacks. The plastic insulation between the components is either squeezed out of position (so a short circuit), or it has not been assembled properly and prevented two metal parts that should be touching from giving a good contact.

Dave · 2016-12-23T13:35:32.740Z

Thanks Robert, what voltage diodes would you recommend? also you recommend a pair of diodes in series? what benefit do you have using two diodes rather than one?

Regards
Dave

Robert.Wall · 2016-12-23T13:50:24.273Z

What voltage do you want? Anything that gives negligible leakage above the working peak that you need for the input should be fine. The YHDC used to be 22 V, the latest are lower.

You need two in series because the “reverse” voltage of a zener diode is ≈ 0.7 V, the same as the forward voltage drop of any diode, and you don’t want it clipping at 0.7 V one way and (say) 15 V the other way. (Any diode will show the zener breakdown characteristic in the ‘reverse’ (blocking) direction - but the voltage at which it happens isn’t controlled. So a Zener diode actually works - passes current - in the “reverse” direction compared to an ordinary rectifier diode.)

[from: Semiconductor Primer, semiconductors 101]

Dave · 2016-12-23T19:30:53.893Z

Hi Robert
I’m trying to get my head around which bias to connect the diodes, wouldn’t you connect them in parallel reversed bias so one conducts on each cycle?

Regards
Dave

Robert.Wall · 2016-12-23T22:36:21.563Z

No! If you did, the pair would clip at 0.7 V. Whichever way round the voltage is, one always conducts at 0.7 V, the other doesn’t get a look-in.

If they are in series but facing opposite ways, one conducts at its zener voltage and the other conducts at 0.7 V. With the polarity to the pair reversed, the first conducts at 0.7 V and the second at its zener voltage. Result: they clip either way at (zener voltage + 0.7 V).

Dave · 2016-12-28T08:48:10.842Z

Thanks Robert, that makes sense as were interested in the reverse bias breaking down at the required voltage.

Regards
Dave

pb66 · 2016-12-28T21:35:22.860Z

8 posts were split to a new topic: Serial-Direct connection between Pi and emonTx shield

pb66 · 2016-12-28T22:09:53.229Z

A post was merged into an existing topic: Serial-Direct connection between Pi and emonTx shield