Maker Pro
Maker Pro

Using an oscilloscope without a probe. Good or bad idea?

Ch33f

May 18, 2015
27
Joined
May 18, 2015
Messages
27
Hi there,

I want to measure current with my oscilloscope. So far, so easy. Just bodge in a little and precise resistor, like 1 Ω or so and measure the voltage across it.

But I don't want to bodge that, I want a nice sollution. Now here is my idea:
I wanted to make a little resistor box e.g. like this: http://www.instructables.com/id/Make-a-Variable-Resister-with-1-Million-Settings/
Only with 4mm banana plugs, so I can use it in a convienient way with my lab power supply.
Additionally to the banana plugs, I wanted to add a bnc connector, so I can hook it up to my oscilloscope too, in order to measure current with it.
Here is my little dilemma: I don't have one of those neat probe to bnc adapters and it's a pain in the butt to find one in germany without having to buy a whole expensive probe.

So, what if I just plug the resistor box into my oscilloscope without a probe in between but directly via a bnc cable?

I am fairly new to the field of oscilloscopes but as far I understand, the compensation circuit in the probe is there to prevent over/undershoot while maintaining a good rise/fall time but I don't know how bad of an idea is it to work without a probe/compensation circuit.

Is that a no-go, with potential to blow my oscilloscope or will it just mean I have less precise readings?

Thanks in advance to anyone taking the time to read this! :)
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
12,890
Joined
Nov 17, 2011
Messages
12,890
So, what if I just plug the resistor box into my oscilloscope without a probe in between but directly via a bnc cable?
This is basically possible. Add a BNC socket to your box and connect it to the scope via a standard BNC cable.
The input resistance of the scope is high enough not to have a noticeable influence on the measurement when the sense resistor is small (a few Ω).
Compensation is not necessary in this setup. Compensation is required for probes which divide the input vltage (e.g. x10, x100) because the resistive divider (probe+scope input) has parasitic capacitances that you need to trim to give a flat frequency response.
Note that a scope's ground is usually connected to protective earth. And even if it is not directly (DC) coupled, there is always some capacitance that gives AC coupling. Therefore you have to set up the measurement carefully to avoid ground loops which wil induce a lot of noise or hum and make your measurement unreliable.
 

Colin Mitchell

Aug 31, 2014
1,416
Joined
Aug 31, 2014
Messages
1,416
Scopes are earthed and if you want to take any readings above earth, you need to remove the earth pin or use an adapter that does not allow the earth to pass through the adapter.
 

Ch33f

May 18, 2015
27
Joined
May 18, 2015
Messages
27
Thanks for these quick answers!
I'll try that out then.

But I have a followup question:
Therefore you have to set up the measurement carefully to avoid ground loops
if you want to take any readings above earth, you need to remove the earth pin

Do these two statements not contradict each other? Or am I misinterpreting the term "above earth"?

From what I understand, according to Harald Knapp I should have the test circuit be floating because oscilloscopes grounds are are connected to earth. But according to Colin Mitchell I should make my oscilloscope float as well, even if the circuit under test is floating...

So what is the best/worst case out of the following scenarios?
  1. Osci. earthed - circuit earthed
  2. Osci. floating - circuit earthed
  3. Osci. earthed - circuit floating
  4. Osci. floating - circuit floating
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
12,890
Joined
Nov 17, 2011
Messages
12,890
Both floating minimizes interference.
One of the two floating is sufficient in a lot of cases, if not most.

I disencourage removing the earth from the scope, it is there for reason: electrical safety.
 

Ch33f

May 18, 2015
27
Joined
May 18, 2015
Messages
27
Thanks a lot! That will help me in my future projects! :)
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
The grounded scope issue has been a bugaboo since it's inception. Precisely for the reason you posted the topic.. "Current Measurement". Because scopes are grounded it necessitates that one end of the current shunt be connected to ground.

Consider this scenario: You're working on a transmitter with a 1000V plate supply. The transmitter's chassis is grounded. You then remove your scope's ground pin so you can measure the voltage drop across a plate damping resistor or any other component located in the high side rail, even a current shunt you installed there. You then attempt to connect the ground leg of the BNC input of the scope to the plate circuit. At this point your scope's ground rail is now sitting at a deadly 1000 Volts and if your fingers are touching the bare metal of the BNC or alligator clips you're now physically making contact with 1000V too!

Most scopes today are 2 channel scopes. Current can be measured in the high side rail by simply measuring the voltage drop across a resistor using the Differential voltage between input(A) and input(B). It's a hell of a lot safer too!

In short I would NEVER recommend removing the scope's ground! :eek:

Chris
 
Top