CT Accuracy: CT-40 versus CT-50

[peppersass]

I'm ready to order a GEM, about 40 CTs and a DB. I had originally planned to buy CT-40s for most of my circuits, but I read a post here that said customers are opting for the CT-50 instead because of the greater accuracy (1% for the CT-50 versus 3% for the CT-40.) But the description for the CT-40 says "Typically 1% on average" and the accuracy chart shows worst case of about +/- 1%. There's no corresponding chart for the CT-50 to compare.

So, are the CT-40s really comparable to the CT-50s as far as accuracy is concerned?

I'm willing to pay more if the CT-50 is materially more accurate. I'm also asking is because the CT-50s are out of stock with no ETA yet and I'm trying to decide if it's worth waiting to order my system.

[sub3marathonman]

So, are the CT-40s really comparable to the CT-50s as far as accuracy is concerned?

I'm willing to pay more if the CT-50 is materially more accurate. I'm also asking is because the CT-50s are out of stock with no ETA yet and I'm trying to decide if it's worth waiting to order my system.

Here is the chart for the CT-40:
http://brultech.com/products/ECM1240/pa ... icro40.htm

Considering that the accuracy is really a range instead of a fixed number, and it was a very similar concern I had, I concluded that ultimately it isn't an issue, at least to me and I like to be accurate.

Instead of the CT-50, a similar CT you could compare to would be a split CT-60. The accuracy chart for that is:
http://www.brultech.com/products/ECM124 ... T%2060.htm

Since if a CT-40 could monitor it, the maximum amperage is 40 for the circuit.

The accuracy with a CT-40 at 30 amps is a maximum of about 0.8% off high, although on average at that point for a typical CT-40 (your specific CT-40 might be slightly more or less accurate) would be maybe 0.05%, which is absolutely spectacular. It could even be slightly off (maybe 1.1%) low. So theoretically, monitoring with many CT-40s, the overall accuracy should approach essentially perfectly accurate.

With the split CT-60, at about 7 times the price, the inaccuracy would probably be about half of the CT-40s. And I might say, at least for 30 amps, the overall accuracy of the CT-40s would surpass the overall accuracy of the split CT-60s, but this applies only at that specific amperage.

I'll defer to Teken to give the definitive answer about accuracy here, , but for me, I would be perfectly happy with the CT-40. (And I just ordered 36 for my project without hesitation.) And beyond accuracy is the practical situation at this point, there are no CT-50s and there apparently is no probability of restocking them any time soon, so the choice becomes moot.

[Teken]

The important thing to remember in all of this is that the main line will be monitored by a 100 / 200 amp CT. Which provides 1% accuracy for the entire system.

I have used all the CT's from Brultech and the only one that has given me a larger margin of error / difference is the micro 30.

I am currently testing out the micro 40 to the micro split 30 to see if I simply have a bad one.

Regardless, of all the above once you start to track your energy consumption and vampire draws. You will be armed with the imperial data to make the changes to save you money, energy, and time.

When compared to other energy monitors on the market even the largest % of error is 2-5 times more accurate than other competing units.

We all look forward to seeing some of the solutions and energy saved once you get your DB / GEM up and running.

FYI: Micro 50 CT's are 3-6 weeks out from being replenished for stock.

Encrypted By: Phoenix Security Solutions

[ben]

We'd be interested to know what you find in terms of accuracy with the Split-30s as they are our most inaccurate CT, they should be within acceptable levels though (2-3%).

[Teken]

We'd be interested to know what you find in terms of accuracy with the Split-30s as they are our most inaccurate CT, they should be within acceptable levels though (2-3%).

I will create a new thread tracking this effort and will publish some charts. Once I can swap out all five split 30 CT's (one at a time) when comparing to the micro 40 that is in place now.

My plans are to take a few known (steady) loads like a light bulb, hair dryer, ceramic heater. I believe this would be a good test showing the lower, mid, and high limits of the sensor and what error rate is observed.

[Teken]

For those interested the testing of micro 40 vs split micro 30 is referenced here: http://www.brultech.com/home/community/ ... f=3&t=1181