measure direction of dc

What's a "pullip"? A pull-up resistor whose grand-parents were tulips?

In the subject you say "DC", now you state 1 kHz. Something doesn't match up here.

If it's actually DC, any multimeter that can measure current at all should do just fine. Since it seems to be a signl line, you'll probably want to use a shunt resistor and an oscilloscipe instead (and possibly a differential probe).

Since you specified the 1 kHz frequency, it appears that this is AC.

What is wrong with a simple current transformer with one turn in the primary and say 1000 turns in the secondary ?

Paul

I assume it's some sort of bidirectional data, and you want to see who is transmitting.

One way is to add a resistor- say a few hundred ohms- in the line- enough so you can see the voltage across it, low enough not to disrupt the logic levels. Position it between the pullup (if it's just at one end) and one of the devices. Look at the voltage across the resistor with a scope - depending on the system and your scope you might need to make a differential amplifier. The magnitude (if single pullup) and/ or direction (if pullups each end) will tell you which device is transmitting.

Paul Burke

Sorry, you are right - i said it in a wrong way.

I have a bidirectional communication on one wire. The bits are sended with 1kHz. With "DC" i wanted say that it is not AC (powernet or so...). And sure, i have Pull-UP resistors ;-).

My goal is, to see who is sending the bits to debug the communication.

Any ideas?

Thanks.

Strictly spoken, this *is* AC, as in "alternating current", even though it's fixed polarity, i.e. the voltage on the line has the same sign, all the time, but the direction of current flow changes, and that's what you have to measure.

Which means my original answer stands: you just need any odd amperemeter. Given the details of the job, it may have to be slightly more advanced than a plain dumb multimeter for good results, but even one of the latter should be able to give you a first impression. For good measure, use an oscilloscope with a differential probe and a shunt resistor (or abuse the signal cable itself as a resistor...)

If it's a pulled-up bus, with active pull-downs: if you put a little bit of series resistance in the wire, watching one end with a 'scope should make things obvious: the local end will pull to a lower voltage than the remote end.

Steve

You add a LED in the circuit across the secondary and you have a visual indicator.

AC by definition means that the current flows first in one direction, then the other. This case could probably be described as pulsed DC.

Regards, Ken Asbury

Put a series resistor (10 or 100 ohms) in the data signal and connect the inputs of a comparator (or opamp) to each end of the series resistor. The output of the opamp will tell you which side of the bus is driving the data signal. You will probably need an opamp with rail-to-rail inputs and a low offset voltage.

Connect the output of the opamp to one channel of an oscilloscope and the data signal to another.

-jm

Which, in this case, is exactly what happens. Remember: the OP is talking about a bi-directional link, here, on which the current direction will indeed change depending on which end of the link is sending data (by sinking current to ground on its pin), and which end is receiving it.

Per Merriam-Webster Online: "alternating current: an electric current that reverses its direction at regularly recurring intervals -- abbreviation AC"

I'll stick with pulsed DC

Ken Asbury

I'll call BS on that one. Don't go to Merriam-Webster for electrical terms, and quote the results to a room full of EEs.

If the current is flowing for a while in one direction and then for a period in the opposite direction, a current transformer would still produce a positive or negative going voltage peak, when the current changes direction.

The advantage with a current transformer is that you do not have to worry about the common mode voltage range and if the signal current is reasonable, a simple Schmit trigger after the transformer should be enough.

Paul

If it's proper AC, I would imagine that you will need a Directional Coupler, which makes things a bit more complicated.

Someone will be along in a few moments to show you a fearsomely simple and effective circuit using 3 op-amps and a handful of passive components (or at least that's what usually happens)... Or failing that, post it to sci.electronics.design

Mike

Suggestion dismissed on the following grounds:

1. Too simple 2. Analog 3. Thinking exceeds dimensions of box

Please don't strip attributions for material you quote.

Did you ever use (or price) a Tektronix current probe for a scope? One wire through a toroid, with mechanism for opening it up for insertion, is the fundamental mechanism.

Yes, of course. Also AC ammeters. Brings back memories of working on Fujitsu POS systems where it used the same technique with toroid transformers for each register to pick off and insert into the signal path.

The point was that the idea scored well because it was so direct, simple, elegant, brilliant rather than the other intrusive & overblown suggestions.

Response dismissed for lack of original question not seen in this newsgroup at this site.