Beiträge von Calamar

Zitat

Thema

Einsatz von Fremdwandlern an Shellies der "EM"-Familie

Der erhebliche Preisunterschied zwischen den Original-Stromwandlern für die "Energy Meter"-Shellies und online angebotenen Stromwandlern haben mich dazu bewogen, ein paar Muster vom Laden des wackeren China-Manns zu bestellen:

smarthome-forum.eu/attachment/40134/

Hier ein Größenvergleich aller Wandler mit den Originaltypen:

Von links nach rechts:

(a) in blau ein Durchsteckwandler, Typ DL--CT08CL5, Nennstrom 80A, 2000/1

(b) in weiß der 50A-Wandler von Allterco mit offenen Drahtenden

(c) in schwarz…

thgoebel

24. März 2023 um 13:43

Unfortunately I don't understand German so my searches are limited to English/Spanish/Catalan in most of the cases. I translated your topic and Its quite informative.

I didn't find any trace of diodes in the CT from Alterco, coil is direclty soldered to the harness's cables. Maybe I'm missing something in the translation.

black

_ Black (Personalizado).jpg

red_ Red (Personalizado).jpg

For me further tests are absurd because right now I don't have the proper AC source/load.

I like the idea to swap all CTs to the 50A version, I'ts more within my working range (2-25A). 120A is too much for most SOHO European appliances.

I assume calibration is a simple linear function (slope) or does it use a LUT for lower currents as well?

I dissasambled a current transformer for the "incompatible" 3EM non-pro version and compared it to a Pro 3EM version. I did the same with the four transformers (Original A,B,C) and my new N.

This was documented and tested february 2024.

Inside

There is no hall sensor for DC or any other electronics, its just the coil. Predictible for a two wire configuration. Same height and width for the ferrite.

You can't tell which one is pro and which one is non pro.

_ Comparison (Personalizado).jpg

Measurements

The non-pro CT is marked as "N" on its plastic shell, all other A,B,C CTs are the originals from the Pro 3EM package.

Impedance is the same (within an acceptable range) for both. Coil seems to have the same configuration: a single coil (no parallel paths) same number of turns (inductance) same wire (resistance).

Inductance

_ Original L (Personalizado).jpg_ Non Pro L (Personalizado).jpg

Resistance

_ Original R (Personalizado).jpg_ Non Pro R (Personalizado).jpg

The only difference is the connector.

The connector used is MOLEX micro-fit 3 (not specified in the CT documentation). Pro version uses 2pin connector (2 wires used). Non Pro uses 4pin connector, only 2 wires used. Unfortunatelly molex micro-fit is a very bad connector to reassemble, especially the female connector.

It is possible if you have two strong <0.1mm thick and about 2mm wide very-rigid metal sheets to push on both sides (left and right) of the connector from the outside.

Modification

I decided to addapt the 4 pin to a 2 pin version until a new 2 pin connector and pins arrives.

To do it you only need to move the red wire from the side to under the black wire.

Warnings:

- In this case the coil will be "reversed" so you can either: reverse the CT when installed, use the software feature to reverse it.

- You need to carve a little the connector flap for a tight connection.

Because you need to destroy part of the connector to fit it into Pro 3EM you don't need to care about disassembling the female pin from it. Just destroy the plastic carefully until the red pin is released.

_ Modiffied 2 (Personalizado).jpg

Test

Go to the Pro 3EM webserver: first calibrate the probe and check the results.

The calibration menu since 1.2 firmware (or maybe prior to that) is not located where documentation states.

You have to go to HOME --> To see full EM data info click here >

Then on the top mini central menu click on Settings

Screen 2.PNG

There you can, among others:

- Reverse the CT measurement direction (if needed)

- Phase to phase calibration the new N current

Screen 3.PNG

Remember to calibrate any CT: ambient should be free of other current sources, both CT should be connected on the same wire and if possible well aligned.

If you did the connector recycling trick (4pin to 2pin): neutral CT should be connected facing the other power-flow direction (reversed).

Then Select the Source and Destination calibration CTs and calibrate.

Screen 4.PNG

In this case I'm testing all probes wired in a single phase resistive load. 8A.

Screen 5.PNG

Conclusion

Both types of CT seems to be the same except for the molex conector and pinout. They share electrical properties, configuration and size to be interchangleable. Any 4pin-to-2pin molex and and Xpin-to-terminals dongle should do the trick to use the CT on any of the shelly power measurements devices. Connector specifications should be available as well. The pro version should always include this dongles no matter the "eco" argument. The instructions in the official documentation to calibrate CTs on Pro 3EM is outdated and points to the wrong direction.

The Current Transformer situation in the Shelly cosmos is rather dissapointing, especially on the "pro" version of its hardware. Lack of stock, lack of specifications, lack of updated documentation, thats not pro at all.

Hi,

I have one of the Pro 3EM and I bought an additional 120A current clamp for the neutral.

Problem is, it comes with 4 pin connector instead of the two pin connector.

Searching in the documention:

States that the correct part number is:

Now I'm trying to buy the correct part but I don't seem able to find a shop where the specific clamp for the pro 3EM is available.

Either:

don't show the connector

there is no connector (bare wire)

its the four pin connector

In the official site there is no variant for the "pro 3EM":

https://www.shelly.com/en-es/products/accessories-and-merchandise#unfiltered

Can you help me find the proprer part for this device at any shop in Europe?

I've got a 3-phase installation.

I want to check several consumptions split over several phases, to check individual consumptions. I would like to use it as three independent channels until I have my loads characterized.

IExample 1 works, It means I could use 3 R loads, or 2S and 1T loads, or any other combination. Setting up voltage input for the specific load you want. In this case if your clamp changes the phase you have to change the voltage input as well.

Example 2 means, you set up voltages sensors and then by software you assign every current clamp to the phase you want as monophasic load. In this case you dont need to change voltage inputs once set up.

Where I can find the manual, mine came with small instruction papper and the same schematic you can find on the side of the device.

Can you refer me to the full manual of the product?

It is possible to use any of the 3EM devices as independent current/power measurement channels connected to the same phase?

As long as the voltage phase matches the current clamp. Of course power factor and composed measures render useless.

for example 1:

channel1 --> R --> voltage to R --> current clamp on desired load.

channel2 --> R --> voltage to R --> current clamp on desired load.

channel3 --> T --> voltage to T --> current clamp on desired load.

+N

for example 2:

Voltage connected to R, S, T

channel1 --> R --> current clamp on desired load --> Software assigned to R

channel2 --> R --> current clamp on desired load --> Software assigned to R

channel3 --> T --> current clamp on desired load --> Software assigned to T

+N

Ideally I would like to associate a current clamp to a certain channel and keep the voltage inputs as regular 3-phase. If not, can I connect each voltage channel to the desired current measure?

Currently is running firmware version 0.10.2-beta1

Available update 1.0.7

The device does not update firmware via link method or "update" button on the webserver. Before briking it by forcing firmware update to the current available. I would like to use an "older" firmware to jump from that beta version as a midstep.

Can someone suggest me an intermediate firmware version to jump from 0.10.2-beta1 please?

There is too much outdated information or not completely coherent answers regarding these questions.

I've selected (bottom) some of the topics and their answers, as well as some of the official documentation to try to answer the following questions, but some remain unanswered (??) or with an incomplete answer.

I would like clarify some FAQ of this product not yet "answered" as of 2023/06 (Firmware 20230510-083155/v1.13.1-gda6f9f2)

What is the minimun polling time of 3EM?

??

How does the 3EM handle non-linear loads i.e: inverter, pc-power supplies etc?

??

What is the accuracy/precision of the measurements? Do they comply with any standard?

pasted-from-clipboard.png

Some topics state 1% accuracy (other people describe it as precision...) but do not clarify if in voltage, current or phase and if its related to nominal values or any other reference value.

EN 50470-1 for reference.

After all most people, IMHO, would be satisfied to know the expected deviation/error in energy mettering per month f.e. in +-kWh, to compare with their bill.

Can 3M handle a 4th current clamp for the neutral conductor and show its data as a normal channel?

It can handle the 4th clamp for sure if its 120A model.

Regarding data collection N is not considered as a fourth channel in the interface.

??

Which is the correct accessory for 3EM the 120A, the 50A or both clamps?

Some topics state that the 50A is not compatible with the 3EM, but the box of the 50A clamp states its an accesory for 3EM.

The following link states that "other transfoerms may destroy the device" but some topics describe other transformers as compatible and secure.

https://www.shelly-support.eu/lexikon/entry/353-3em-calibration-of-the-n-clamp/

If the 50A clamp is compatible, which is the reference, part number, specifications for the 4 pin connector? (50A comes with bare wire)

??

How can the neutral current clamp work ("uncalibrated") as an accessory but the provided current clamps are "factory calibrated" and are not interchangable? Does the accesory come with a calibration file?

3EM provides a current calibration system in its webserver. Instructions are in the LEXICON of this site.

pasted-from-clipboard.png

Is the energy (power*time) data logging (In/out) corrected or you still need a third party software to compute it?

??

Related topics with no "clear answers":

Since shelly output is a relay and schematic is depicted in:

as

[Blockierte Grafik: https://kb.shelly.cloud/__attachments/35094529/Plus-1-internal-schematics.png?inst-v=fba0f242-f9c8-47aa-85fd-3758357f05b6]

Can be connected as follows:

Powered by AC phase voltage, using phase voltage as switch.

Shorting (closing) a DC circuit, being DC from lets say 1V (<0V) to rated DC voltage.

A) FLOATING DC:

pasted-from-clipboard.png

B) GROUNDED DC:

pasted-from-clipboard.png

I've consulted some diagrams available of the "1" version (about 12 wiring diagram combinations available) but those are not among them.

Existing Shelly EM, I extend my question, is not possible because Shelly EM exists (so they won't work on that feature to avoid product overlap), is a work in progress?

AFAIK all you need is to save a value in a non volatile register (same as other configuration parameters) and a couple of changes in the UI: "reset data" button, "activate/deactivate feature" slider.

I've noticed that, when the plug is moved to another location (i.e. unplugged-plugged) or there is a grid or power outage no historic data remains for energy consumption or power. I've seen no configuration avaiable regarding this issue in the webserver.

Firmware: 20230503-101129/v1.13.0-g9aed950

I've noticed another topic regarding this:

But its 3 years old. Any update in this regards?

Is there a way to save that data? Of course without clouding or third party software...

I opened the question, thank you.

What is the viability, or possibilities to implement WPA-Ent network compatibility for your products?

Now it is possible to use a certificate, but not a specific username/password for the logging.

Ty.