clamp meter load

[ILCML]

When measuring one leg of a 3 phase (coming out of a phase adder, fyi) with a clamp meter (FLUKE 336) there should be no noticable effect on the current or voltage whatsoever, correct?

 

[ScottyUK]

The burden of the instrument will be very low, so no there should be no noticeable effect under normal circumstances. Why do you ask - do you think you are you seeing an effect?



----------------------------------

Sometimes I only open my mouth to swap feet...

 

[ILCML]

I was told by a mfr. that the reason for current fluctuation was due to the clamp meter creating an electromagnetic drain on the current.....I couldn't imagine it being significant, so I thought I'd do a sanity check.

thanks...

 

[CJCPE]

One possible reason for fluctuating or inaccurate reading is the clamp not being tightly closed or the faces where the clamp closes not being clean.

 

[aolalde]

On power circuits, clamping on ammeters do not disturb the circuit parameters. May be it will do for a circuit drawing milli or microamperes.

 

[Skogsgurra]

Actually, a circuit with heavy current and very low impedance can be noticably influenced by the clamp.

An example where this effect can be readily observed is when you have short parallel cables - like what you sometimes find from a power transformer secondary to the bus-bars in the distribution system.

The resistance and impedance of each cable is very low and as long as nothing is clamped on, the current will be evenly divided between the cables. If you measure current in one cable and multiply with total number of cables, you will see that the product is a little less than the total current. The reason is that the clamp increases impedance of the cable measured so that current is decreased somewhat. The two other cables (if you have three parallel) then carry a little more. The effect is hardly noticable - but it does exist.

Circuits having milliamp current usually have a high impedance already and they will not be affected by a current clamp. The field around such a conductor is also very low and does not create any significant counter-EMF in a core around it.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[waross]

In support of Skogsgurra;
The iron in the current clamp will be creating a wireless reactor. These were unwound cores through which the cables of a transformer were passed to raise the effective reactance of a transformer. If only one cable passes through the core or the clamp meter it will raise the impedance of that cable slightly. It's not the burden, it's the iron encircling the magnetic field of the conductor that causes the effect.
respectfully

 

[jraef]

I'll attest to that as well. I have seen it where there are CTs going to a meter and I clamp on my ammeter, you can see the current drop on the other meter ever so slightly under some circumstances. And what CJCPE said is absolutely correct. If the clamp can't fully close tightly, as is often the case when you have multiple conductors or very limited access, the burden jumps dramatically.

JRaef.com
"Engineers like to solve problems. If there are no problems handily available, they will create their own problems." Scott Adams
For the best use of Eng-Tips, please click here -> faq731-376

 

[electricpete]

Although I've not witnessed it, I can believe skogsgurra's scenario since the current in those circuits such as parallel branches would be very sensitive to small variations in impedance.

But a CT? That would be unusual indeed. The CT is like a current source relatively insensitive to change in impedance of the driven load provided you don't push the CT into saturation. It doesn't sound right to me.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Skogsgurra]

To make the scenario a little bit clearer.

A CT (or a clamp-on ammeter) has a secondary winding that feeds a burden. If the burden is low ohms (which it usually is) the core's flux excursions needed to induce necessary voltage is quite low. If the burden is higher, which it sometimes is, the voltage is higher. Strictly proportional to burden.

This voltage is reflected back to primary and produces a counter-EMF in the primary winding, which - if low impedance - will react in such a way as to reduce current. In the three parallel short cables scenario, the effect can be readily observed.

But I do not subscribe to the "not closed yaws" theory. If yaws are not closed, not much flux is developed and the effect is not possible to see.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[waross]

Hi Gunnar;
Remember that the meter core encircleing a power conductor also comprises an inductor with one turn. You have encircled a conductor with a magnetic material and eddy currents will result. Admitted, the laminations reduce the eddy currents, but the inductance remains. This will also have the effect of increasing the impedance regardless of burdens and even without secondary turns.
I believe that the inductive affect will in most cases be several times greater than the burden effect.
Respectfully

 

[Skogsgurra]

Yes Bill, I am fully aware of the single conductor = one turn thing.

And, to take that a bit further, it is the priary winding in a transformer. The secondary winding is usually connected to a low ohm burden, where the measurement is made - be it a relay coil, a meter movement or an electronic amplifier - as is usual in clamp-on meters.

If the secondary were shorted (burden = zero ohms) it would be like shorting the secondary of an ordinary transformer - no counter-EMF and lots of primary current. In other words; no inductor.

That is exactly what happens when you short a CT. No voltage drop across it. It is only when you have a burden where secondary current develops a voltage that you notice the presence of the core. So you can, in effect, have a core circling a conductor without having any inductance in it at all.

It is quite revealing to do a simple experiment: connect a DMM with AC mV range to two points a couple of inches apart on a current carrying conductor. Then put a clamp-on ammeter between them. If you have enough current (abt 10 amps), then you will see voltage increase. Sometimes the voltage is dependent on what current range (different burden) you select.



Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[ScottyUK]

Gunnar,

Have you ever tried that experiment using the hall Effect type current clamps which have a bucking winding to drive the core flux to zero? Curious as to the experimental results if you have.


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[electricpete]

There have been a lot of scenario's described.

I agree there is some very small impedance associated with the primary of a clamp-on ammeter. Since the secondary of the clampon feeds a voltage circuit (very low current), the impedance is simply the same impedance we would get from an iron donut.

A set of parallel cables would have individual cable current unusually sensitive to even small impedances.

So in clamping an ammeter directly onto one of several parallel cables, it seems very plausible that the division of current among those multiple cables might be sensitive to that very small impedacne.

BUT the scenario described on 19 Jan 07 18:02 was a CT whose secondary was feeding a meter. It was claimed that clamping an ammeter onto the secondary circuit affected the reading of the meter and if I follow skogs response the reason would have been that the added impedance of the clampon affected the primary circuit of the CT.

I have two problems with that:

#1 - This impedance of the CT which was very small to begin with is reduced by a further factor of CT turns ratio squared in going from the secondary to the primary of the CT. Very small times something like (1/20)^2 will be ridiculously small.
#2 - The nature of the load on the primary of th CT in this scenario was not described but it would not logically be one of a set of parallel cables because we have a permanent meter on the secondary of the CT (you explain to me why anyone wants to monitor current in individual cables of a parallel cable configuation). Maybe the poster can clarify the primary in thise case, but you can call me skeptical.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Skogsgurra]

Hi Pete,

I do just that. Typicall several times every month. The reason is simply that if you have a set of parallel cables (which you have when a 240 mm2 or 300 mm2 cable is not enough) then the clamp cannot "take" all cables. You then have to measure current in one cable and multiply by number of cables.

It is when you do that that you can see the effect of current "avoiding" the cable you measure.Sorry, I am on my way to Indy now. Staying overnight in Gothenburg. So I do not have acess to measurements I made.

If, to make things very simple, we have a CT with 500:1 ratio and a burden that has 1 V voltage drop. Then you have no less than 0.002 V drop across the CT. And that means a noticable current difference in a 5 m 240 mm2 conductor paralleled to other conductors. Go and try it! It is a fact.

Some of the older clamps were real bad in this respect. The newer, electronic ones seem to have less power consumption in the measuring circuits.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[electricpete]

skogs - I think you are missing my point.

"I do just that."

You have a permanently installed meter reading individual cables of a parallel set? I doubt it. I think you're saying you put a clamp-on meter on individual cables. But jraef said this particular CT feeds a permanent meter. My point was that the load on the primary of this CT feeding a meter is not likely one of several parallel cables. That was my point #2 above.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

maybe jraef can tell us about what was on the primary of his ct?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Skogsgurra]

Pete, we agree that a clamp on a cable that feeds a load has next to no effect on the current.

The discussion then took another turn and was more or less about the possibility that a CT an influence current at all. It can. And that has caused some confusion with my customers in the measurements I mentioned.

So, what I said is that that measurement situation sometimes occurs. And I think it is good to have some knowledge about it.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[electricpete]

I respect your opinion and expertise completely, as I also do jraef and many others.

Maybe somewhere along the line there was confusion on my part or another's part about who was talking to who.

I didn't disagree with any of your comments gunnar except in the context it appeared you were defending the scenario posed by jraef. You directed your comments to me, but all of my comments are specifically to question the scenario of jraef.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[GregLocock]

Um, if we are talking about a power circuit, to have an appreciable effect on the circuit, one would have to absorb a lot of power in the instrument.

Most clamp on devices seem notably devoid of cooling fans, or heat sinks.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.