Are there any aftermarket CTs that will work?

[shortwh]

I have some hybrid inverters and they came with CTs that I don't need but when I tried them they calibrated at low power but overwhelmed the GEM at higher loads. They stated they were voltage based at 333mV but that must not have been correct. Anyway, has anyone had success with different CTs. I need several and would really like 200A coils that are current based around 20mA at max load. Could I add a resistor to these voltage coils I have and quarter the voltage with a 4k inline resistor? Would that work? Would it cause errors on high or low sides? I wish Brultech had Micro-200 coils that were current based.

Current based because I am using CAT5 to extend these coils and I only have 12pairs (3 CAT5) cables in the ground. If I use voltage based it uses 4 wires(2pairs) per monitored device(L1/L2) where as the current only use 1pair per device.

[ben]

Do you have the spec sheet on the 333mV CTs?

I ran the voltage/current-based CT question past Paul. He wasn't sure what you meant by the current CTs using less lines, you should be able to combine L1/L2 to sum the 2 CTs on 2 wires.

[shortwh]

Maybe I misunderstood originally. I used Type-A CT (Split-200) and I carried four wires back to GEM. When I hooked up Micro-100 I only carry 2 wires back to GEM. Is there a way to use the Split-200 with only 2 wires back to GEM?

[ben]

Maybe I misunderstood originally. I used Type-A CT (Split-200) and I carried four wires back to GEM. When I hooked up Micro-100 I only carry 2 wires back to GEM. Is there a way to use the Split-200 with only 2 wires back to GEM?

Ah, gotcha. Yeah you can splice them on 2 wires at the CT end. Use the bottom and top ports (1,4) instead of 1,2/3,4.

[shortwh]

(And I see that you realized and I forgot to mention that L1/L2 where both being monitored so there are 2xSplit200 and 2xMicro100 coils involved in the above post)

So, that makes sense. I wish I had asked you guys years ago.

So where is the 1000 ohm resister for type A and the 6.8ohm for Type-B? in the CT or on the board?
Do you have a little mocked up circuit diagram of those resistors and there relationship to the pin1,2,3,4 of each channel?

[ben]

So, that makes sense. I wish I had asked you guys years ago.

So where is the 1000 ohm resister for type A and the 6.8ohm for Type-B? in the CT or on the board?
Do you have a little mocked up circuit diagram of those resistors and there relationship to the pin1,2,3,4 of each channel?

I think the quick guide here explains it best: https://www.brultech.com/software/files ... id_QGD.pdf

Across 2,3 has a built-in burden (4.99ohm). The others are unburdened.

[shortwh]

I pulled my numbers from the GEM-MAN_ver4_1.pdf page 45.

Input Impedance:  1,000 ohms for Type A CT connection. Terminals 1&2 and 3&4
 6.8 ohm (burden resistor) for Type B CT connection. Terminals 2&3

So back to the original question. The CTs which seemed to be overpowering the GEM. Can they be used if an extra resistor is used? They claimed 0.333mV at 300A and 100mA at 300A which seems strange and they had 20ohms printed on it. They calibrated at 748w but their CT Type / Range values where 90/5. When I tried jumping to 10000w they sent the GEM into all manner of strange readings on all sensors. I quickly removed load and all went back to normal readings but they are the only CTs I have on hand and didn't want to spend another $180 on 4xSplit200s.

[ben]

I pulled my numbers from the GEM-MAN_ver4_1.pdf page 45.

Input Impedance:  1,000 ohms for Type A CT connection. Terminals 1&2 and 3&4
 6.8 ohm (burden resistor) for Type B CT connection. Terminals 2&3

So back to the original question. The CTs which seemed to be overpowering the GEM. Can they be used if an extra resistor is used? They claimed 0.333mV at 300A and 100mA at 300A which seems strange and they had 20ohms printed on it. They calibrated at 748w but their CT Type / Range values where 90/5. When I tried jumping to 10000w they sent the GEM into all manner of strange readings on all sensors. I quickly removed load and all went back to normal readings but they are the only CTs I have on hand and didn't want to spend another $180 on 4xSplit200s.

If you can send me the spec sheet I can pass it onto Paul and he should be able to figure out a way to wire them up.

EDIT: Even just the make/model I should be able to pull enough info for him.

[shortwh]

I'm sure you already realized that I was confused by multiple wrong sources on the internet about these CTs. CMO324

I disassembled them and found they have a 6.8V TVS to limit the output voltage in case someone fails to connect something to output (or like I did at first assume they are internally loaded). I did many calculations with measurements to figure out the number of secondary turns. None of that worked out to match the physical measurements. So I built up a known current "Load" and randomly choose burden resistors until I got the voltage of a single CT down to 333mV at a calculated maximum amperage. Turns out these need a 6.8ohm burden resister(surprise, surprise. I wonder were I saw that number before.) However, they produce too much current to use the embedded burden resistor on GEM when I use two of them on the same channel. I think that I should be able to add external burden resistor of 6.8ohm to each and then connect them like TypeA CTs. I didn't have the perfect fit resistor so those are on order.

So before I try again to calibrate these in GEM does my plan make sense? Do I have the logic correct? Also, after having been through this much effort it appears that many different CTs should work, it just comes down to getting them setup correctly and how bad is there accuracy over the full range. If I wasn't so cheap I would just buy the Split200s and remove my headache but $180(4 Split200 shipped) verses $8(20 burden resistors on Amazon) was too great a divide as I had 4 of these unused 300A CTs already.

[ben]

I'm sure you already realized that I was confused by multiple wrong sources on the internet about these CTs. CMO324

I disassembled them and found they have a 6.8V TVS to limit the output voltage in case someone fails to connect something to output (or like I did at first assume they are internally loaded). I did many calculations with measurements to figure out the number of secondary turns. None of that worked out to match the physical measurements. So I built up a known current "Load" and randomly choose burden resistors until I got the voltage of a single CT down to 333mV at a calculated maximum amperage. Turns out these need a 6.8ohm burden resister(surprise, surprise. I wonder were I saw that number before.) However, they produce too much current to use the embedded burden resistor on GEM when I use two of them on the same channel. I think that I should be able to add external burden resistor of 6.8ohm to each and then connect them like TypeA CTs. I didn't have the perfect fit resistor so those are on order.

So before I try again to calibrate these in GEM does my plan make sense? Do I have the logic correct? Also, after having been through this much effort it appears that many different CTs should work, it just comes down to getting them setup correctly and how bad is there accuracy over the full range. If I wasn't so cheap I would just buy the Split200s and remove my headache but $180(4 Split200 shipped) verses $8(20 burden resistors on Amazon) was too great a divide as I had 4 of these unused 300A CTs already.

Yeah, everything there sounds reasonable.