# Current sense transormer to detect current surge in DC link?

M

#### mook Johnson

Jan 1, 1970
0
I know current sense transformers are use for measuring AC currents. I'd
like to use one to the measure a DC link at 1A that could have a rapid surge
to 10A (like in 1uS).

I only want to detect the surge and not read the SC level. I'd also like
for it to be small since I'm only interedted in the surge. It doesn't even
have to be accurate as I only need to detect the presence of the surge.

Would a small cored ferrite transformer something like 0.700"OD or smaller
saturate from the DC current and not detect the high frequency surge?

D

#### D from BC

Jan 1, 1970
0
I know current sense transformers are use for measuring AC currents. I'd
like to use one to the measure a DC link at 1A that could have a rapid surge
to 10A (like in 1uS).

I only want to detect the surge and not read the SC level. I'd also like
for it to be small since I'm only interedted in the surge. It doesn't even
have to be accurate as I only need to detect the presence of the surge.

Would a small cored ferrite transformer something like 0.700"OD or smaller
saturate from the DC current and not detect the high frequency surge?

I might get this wrong but here goes...

Ferrite toroids: ~5000 gauss sat level

Field intensity H = (0.4*pi*NI)/l amps,turns,cm

Perm. ui = B/H gauss,oersteds

Ferrite perm ur: Ugh! There's all different types!

Oh..I'll just make one up

Let's say ur = 3000

Free space perm u = 1.257 x 10E-6 Henrys/m

Since ur = ui/uo

ui = uo*ur

Hsat = Bsat/ui

Then solve for N (turns) using 1A and the magnetic pathlength for a
0.7 OD toroid, you should be able to solve for the limit for turns to
avoid saturation.

There's also a crapload of technical detail that can mess up the math.
So even if I got it right, it's probably still wrong.

OR.. You can skip all that math and just look at a bunch of
datasheets.
The inductance and current are often spec'd.

D from BC
British Columbia

R

#### Robert Baer

Jan 1, 1970
0
mook said:
I know current sense transformers are use for measuring AC currents. I'd
like to use one to the measure a DC link at 1A that could have a rapid surge
to 10A (like in 1uS).

I only want to detect the surge and not read the SC level. I'd also like
for it to be small since I'm only interedted in the surge. It doesn't even
have to be accurate as I only need to detect the presence of the surge.

Would a small cored ferrite transformer something like 0.700"OD or smaller
saturate from the DC current and not detect the high frequency surge?
The "surge" as you call it is an AC signal.
THe current transformer core must not saturate in the presence of the
DC current; so a dinky ferrite transformer will not do.
Ferrites are generally for high frequency - say from 10KC to 100MC,
where laminated iron cores aer gneerally for low frequencies - say from
10Hz to 10Kc.

B

#### Bruce Varley

Jan 1, 1970
0
mook Johnson said:
I know current sense transformers are use for measuring AC currents. I'd
like to use one to the measure a DC link at 1A that could have a rapid
surge to 10A (like in 1uS).

I only want to detect the surge and not read the SC level. I'd also like
for it to be small since I'm only interedted in the surge. It doesn't
even have to be accurate as I only need to detect the presence of the
surge.

Would a small cored ferrite transformer something like 0.700"OD or smaller
saturate from the DC current and not detect the high frequency surge?
From what you've described, it sounds like you should consider air cored.
Depending on your accuracy requirement, there might still be some design
involved, but you don't need to worry about magnetic material behaviour,
meaning that you can ignore saturation caused by the DC.

W

#### Winfield

Jan 1, 1970
0
Robert said:
The "surge" as you call it is an AC signal.
The current transformer core must not saturate in the presence of the
DC current; so a dinky ferrite transformer will not do.
Ferrites are generally for high frequency - say from 10KC to 100MC,
where laminated iron cores are generally for low frequencies - say from
10Hz to 10Kc.

Well, I'd say this all depends on Mook's meaning
of "small", and your meaning of "dinky". A single
turn carry 1A would not saturate most small cores
I'm thinking of, but then, what does "small" mean
to me? Anyway, one turn through the middle of a
"small" toroid core, 100 turns wrapped around it,
and a 27-ohm load resistor, will give a 2.7-volt
output pulse when Mook's 10A current spike arrives.

Replies
14
Views
1K
Replies
5
Views
954
Replies
0
Views
990
Replies
3
Views
2K
Replies
3
Views
2K