Base stations, CTs, North America

Getting Started
1

Hello all - researching this site in preparation for buying a system. Newbie questions:

I’m thinking emonPi w/AC-AC sensor; emonTx w/US AC, 6 CTs, and 5 temp sensors.
Thus, I can baseline energy usage – whole house, A/C unit, and perhaps a couple of fridges and a water heater. (and temps in direct sunlight, shade, attic, indoors, etc) Then as I make energy efficiency improvements, I can track ROI. Perhaps add a PV generation at some point. Reasonable?

Base stations

  1. The emonBase is a “subset” of the emonPi, yes?
  2. The emonPi “bundle” and emonPi-Solar “bundle” are the same, except for number of bundled sensors, yes? So…same soft/firmware, just deployed for different applications? I only wish to baseline power usage and then measure ROI as I improve energy efficiency, but may eventually move toward PV generation.

CTs for USA
My main 240VAC feed wires from utility meter to breaker box/load centre are about .67” (17mm) in diameter. So, I need CT’s from the sources listed in the “EmonTx-in-North-America” link
https://openenergymonitor.org/emon/buildingblocks/EmonTx-in-North-America

  1. Anyone have experience or suggestions (ease of use, accuracy vs cost/value) between the Magnelab, Wattcore, or Sentran?
  2. Hmmm…could I gently file 2mm off each internal face of the CT100’s plastic housing to “make it fit” around my wires?
    CT Accuracy
    I CAN use the CT100s for monitoring other circuits – Air Conditioner, fridge, water heater for example. However, I understand that CT accuracy decreases near low end of CT range. So:
  3. To monitor 30-Amp Air Conditioner circuits, accuracy improves with a 50-Amp CT, yes?
  4. To monitor 20-amp circuits, perhaps use 30-amp CTs.
  5. Or am I being overly concerned here?

Thanks much!

2

Basic concept: Yes, reasonable.

Base stations - 1: Or the emonPi is a superset of the emonBase, having half an emonTx inside the box as well.
Base stations - 2: Basically, yes.

CTs: I’d definitely NOT advise that. One touch on the ferrite core and it might snap. Result: one dead CT.

The accuracy decreases at the bottom of the range for all CTs. The more you spend (i.e. "revenue’ grade), the less it’s likely to degrade.

3, 4 & 5. Probably yes to all 3. The YHDC SCT-013-000 starts to lose accuracy significantly below about 1% current (it’s only specified down to 10%) and part of that inaccuracy is due to the limited resolution of the ADC in the processor. If you’re not trying to measure much below 1 A, the errors should be tolerable. I can’t comment about the CTs listed on the N. America page, I haven’t tested any of them. You’ll need to check the manufacturer’s or factor’s websites for specs and prices.

3

Hello Steve,

I have the same 240 Volt split-phase system you’ve got. (I’m in Oklahoma)
I use CTs from both Magnelab and Wattcore. There doesn’t seem to be any appreciable difference in operation, but there is a definite difference in cost. (Magnelab being the more expensive of the two) The only downside , if you could call it that, is the Wattcore units sometimes have a significant lead time from order to delivery (3-4 weeks)

I’ve no experience with the Sentran units, as they are a recent addition to the list of suitable sustitutes for the Shop-sold CT.

I managed to make my Wattcore WC-1 CTs fit my 4/0 feeders by very carefully trimming the plastic housing with an X-Acto knife. But as Robert said in his post above, the ferrite material is very brittle, which makes it easy to render your CT useless unless great care is exercised. No way would I use a file. X-Acto knife, yes, but file? definitely not!

4

Brultech is a company I remember from the old forum.

Maybe they have something for monitoring the individual circuits. Not as fun as the OEM solutions but it might help. http://www.brultech.com

http://www.brultech.com/store/category.php?id_category=38
http://www.brultech.com/store/product.php?id_product=86

5

I believe it’s their GEM (Green Eye Monitor) you’re thinking of Jon.

6

Thanks Robert!..for confirming my suspicions on both the base stations and the danger of messing with the CT100 form factor. (but…you know…I’m a male…so may try it anyway)

7

Thanks much Bill,
Exactly what I was looking for. If they’re about the same performance, I’ll probably try the Wattcore (and patience waiting for their arrival).

8

Thanks Jon,
but confused about your suggestion of the Bruitech or GEM products to monitor individual circuits…isn’t the emonTX perfect for this?
Cheers,
Steve

9

The Brultech GEM (Brultech is the company name, GEM the product name) has 32 monitoring channels, the emonTX has four.

10

you could always try the ct from ted 1001 or 5000 you can buy those for 20$ ( they come in pair ) on ebay occasionally . i never really tried them connected to emontx, as i just usually hack the mtu and attached an arduino to it… later this fall i plan to take my ted1001 hack and apply it to esp8266 module and see how that works,
example ;
http://www.theenergydetective.com/set-of-current-transformers-cts.html

11

Thanks stephen,
I already pulled the trigger on the Magnelab SCT-0750-100’s. Now just waiting for the emonPi, emonTX, and other sensors to arrive.
Does anyone know what modifications to code must be made to support these MagneLab CTs?

12

The CT-100 sensor from the emon store says its output is 0-50ma.
My Magnelab SCT-0750-100 has a burden resistor, and so .333V at rated current.
Damn…I think I should have purchased the SCT-0750-000, with no burden resistor, which means it has the Zener diode to limit voltage output to 22V, and so has a current output.
I thought the “-000” was just a generic number, and the -100 was for 100Amps, which is what I wanted.
Did I just waste $100?

13

If you have the voltage output CT, then you also need to remove the burden resistor. If you go to Resources > Building Blocks, you’ll find a page there that details how to calculate the calibration constant.

You might not have totally “wasted” your money, but you will have much poorer resolution, which basically means that the lower limit on the current that you can accurately measure will be raised by a factor of about 3 (normally, we design for 1.1 V rms across the burden resistor). You will set the calibration - effectively a software sensitivity control - to give the correct numbers out.

14

You could try asking them if they’d exchange the CTs you purchased for the ones you need.
At least that way it’s just the shipping cost vice the total for two more CTs.

15

Now that I’ve resolved my “rectal-cranial inversion” issue :smirk:, I’ve returned the -100’s for a full refund (-$6 for shipping), gave the seller a 5-star rating for accommodating me, and ordered the -000 units from AIM. There are no -000’s available on Amazon. Thanks much for all of your help guys! Awaiting my eMon stuff to make its way across the big water.
Cheers