Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Using an oscilloscope (for changing current)

Status
Not open for further replies.

malc9141

Member
(I've asked this on Chat but it doesn't seem to get me anywhere)

I have a fast pulse circuit which actuates a diesel fuel injection system. We built the circuit ourselves and it works well. It activates a small coil in the injector to lift a small plate. The plate is magnetic.

I am not an electronics expert - quite the opposite.
But I want to know more - what the circuit is really doing, compared to what we designed it to do.

Question 1
Supply at 15 v. Large current available but (of course) the current is determined by the injectors impedance, which we know. So if I connect a side line from +ve to the Channel 1 and also connect same to the Ext Trigger, I get a picture of build-up & decay of voltage vs time. True?

Question 2:
How do I see what current build-up and decay is? I can't believe I just attach the +ve to Channel 2 via known resistor, and use Ohms Law.
I am also interested in how much reverse voltage I get when the pulse ends and the field collapses, but that's in Q1, I think.

Thank you!

I have just read the link to Oscilloscopes but current isn't mentioned. Is it possible that the current is a perfect reflection of the voltage picture? (I imagined that with the events at switch off, collapsing field etc, induced voltage would oppose the flowing current).

In any case, if I observe change on the screen in Channel 2, with a known resistor in the circuit, would that be pictorially similar to the channel 1 voltage?
 
Last edited:
A current probe is a good option. The low cost probes are "AC" and the $1000 probes will read DC also. You don't need the DC probe. The probe encircles the wire, does not touch. You can get on of these on e-bay.

You are probably driving a coil so the current will not match the voltage.

current_probe.jpg
 
Thanks v much. Just clarify for me.....
The probe tip goes "around" the + action? The blunt end goes into Channel 2. But how am I charting i against time? Does the probe have some factor written on it that allows mew to calibrate?

(I had a helpful man with this last year but he's emigrated to a land of opportunity - Scotland)
 
Theres a good explanation on You Tube (now, thanks to you) I know a bit about what I'm after. I'll see how I go.
 
Question 1 true.

Question 2 ; yes you can measure current by placing a resistor in series with the coil. The value of the resistor should be such that the volt drop across it is less than say 0.3 to 0.5 volt for your supply of 15 volt.
The use of a current probe as suggested is the most elegant solution. In the picture, there is a slide button which opens the magnetic circuit; you put the wire in the slot, and close the cover. The magnetic parts are highly polished and need to be kept really clean. If you buy one of these, get a DC to 50 mHz unit if you can. Just put the wire into the probe and the design will be calibrated at a certain Y amplifier sensitivity. if you put the wire through the probe in the opposite direction, the polarity of the display becomes the converse. When the pulse ends and the field collapses, the waveform will show what is happening. Its not magic; this stuff has been around since the 1960's. (Tektronics 114? and 120?)
hope this helps a bit more.
 
Rumpfy
This is so helpful. Thank you so much! Thank you for reading what I asked (eg 1 & 2). 'course, you would be an Aussie (I'm in Sheffield but the family is in MEL)
Malc (who is an engineer but not an electrical one)
 
In this kind of situation i use the shunt in my multimeter, setup your meter as if you were reading dc current in the coil, then connect your 'scope accross the meter.
Your meter wont read right but the 'scope will.
 
With a two channel scope and a shunt resistor there are several things that can go wrong.
1)The scope "ground" may short out the shunt. The resistor should be on the cold side of the load.
2)There can be bad effects with two probes. One ground to measure voltage and the other to measure current. But there is only one ground as the scope see it. Again watch for current in the ground wires to the scope.
 
ronsimpson
OK. I will be sure to to do the Tests, one at a time as well as two. And believe the "one," if in doubt.

I couldn't get access to the workshop yesterday.
I have a little commercial voltage generator (call it VG) which also shows i so I can drive the system with a low safe voltage . Have input +ve > my circuit > multimeter > oscilloscope BNB (or whatever the link is called). And the O v of the VG to the oscilloscope chassis (??)
Or should I put in a simple R rather than the convenient multimeter??

I can see the i for a given v, and so calibrate the oscilloscope - I think!
 
Last edited:
The liklihood is that the coil is grounded and depending on how the coil is made, you may have trouble disconnecting the earthed end of the coil. A lot of auto stuff is made with the earth end of the coil welded to the frame of the device and when bolted on to the vehicle, the earth connection is made. So just be aware. Ronsimpson is also suggesting to take care.
Anyway, these tests could probably be done on the bench away from the vehicle. You might wanting to look at the current AND voltage waveforms simultaneously in order to check the sustaining voltage and other switching characteristics so care with the setup will be necessary. These kinds of tests are more likely required at the design/device selection stage, and its not clear whereabouts you are at..
 
OK. The coil is not earthed to the cylinder head. This is a single cylinder experimental engine we have designed & built.
The circuit we built (thanks to a superb elderly builder of amplifiers) works well - almost unbelievably. The fuel switching is rather fast (!!!) because bleeds are opened or closed by fluid (the diesel) pressure > 1000 atm (you inflate a car tyre to + 1 atm). The circuit needs to be on/off pretty quick, also. This is a mechanical device moving in microsecs ~400 - 2000 us).

The + and - to/from the injector (=coil) goes to/from the circuit board. This can be powered by my voltage regulator (so I know V and i in this circuit).

I presume I (get a square wave from the probe then) attach the probe hook onto the side of the + drive to the coil, and the side arm of the probe onto the auto trigger, plug probe end into Channel 2 (see below).

But I actually need an R between the probe hook and the + drive to the coil. *
Surely its across the Multimeter, switched to read voltage???? That gives the unspecified R - does this make sense?

As for testing V and i simultaneously, do I "sense" (=probe = touch) the + drive to the coil with a shielded wire (wire at one end, connector at the other) attached to Channel 1?

Sorry this is bit wordy.

Malc
* Rumpfy, you say in R in series with coil. "Don't drop voltage much". This would mean disconnect return from coil. But how is the circuit complete? Do I join the 0V on the circuit board to the oscilloscope chassis somehow? Not sure about this bit.

I see your current probes are induction devices - I have a probe that hooks onto a wire. Not the same thing.
 
Last edited:
see attached drg (microsoft word format.
 

Attachments

  • How to measure current with a CRO.doc
    22 KB · Views: 285
I dont mean to sound like a lurker here, however if you test the system on the bench which is a good idea you need to replicate the circuit within the ecu, any protection or snubbers within the ecu could affect the waveform and even operting time of the injector valves, only by uS, however at 7k rpm thats a few degrees.
 
I'll try to master this. Comment appreciated. I'm not trying to mimic timing (of injection). The story is: I found the impedance & inductance of the injector, and believe those values.
That gives an immediate tau for the coil.
Then I observed with CRO the rise and fall in V against time - which also made sense.

But with the CRO, the rise in i, tho' looking qualitatively as expected, gave a very different value for tau.

So (see my 1st post) I wanted to repeat this. Its the change in i that I'm after. So I would use the actual circuit entirely.
Its where to put this R that confused me (eg what is earth, ground, 0 in this set-up?).

NB I still think when using a multimeter as R, it should be on V setting (gives high R) not current setting: do not understand "use current"
 
so long as your 'scope isnt grounded you can put the r in the + or -.
In ye olde 'phone exchanges they used to put caps accross mechanical switch windings to speed them up.
 
Dr peppers comment about using the multimeter was to use the CURRENT ranges. On the current range, there is a low value resistor shunting the multimeter leads. The purpose of this is to shunt all the current except that required for the meter movement. The only problem with his suggestion is that without some calculations, you dont know the exact value of resistance and therefore your CRO 'current' range will be uncalibrated. If you use a known resistor, your CRO 'voltage range' is converted to a known 'current range'.
You are trying to do some theoretical calculations and that is fine. With inductive circuits, if you try to get them to turn off quickly, they will produce a high back-emf which needs to be considered. The generation of this high back emf results from the self capacitance of the coil. The energy stored in the inductance of the coil (0.5 x L x I2) has to be dissipated to get the coil current to zero. With no 'snubbing', the energy will transfer to the self capacitance of the coil and it will do this at the resonant frequency of the coil inductance and the associated self capacitance. The energy will be stored as an electric field in this self capacitance, and the voltage will be (0.5 x C x V2) and the frequency of the energy transfer will be f= 1/2 x pi xSQRT(LC)
You are possibly looking at this subject just as a simple current fall in an inductive circuit, but it is actually a resonant circuit and it RINGS. You should see ringing and the ringing will decay. If you dont want ringing or a high back emf, then the trick is to force the current to decay through a reverse biased diode. This will force the back-emf to continue the current flow in the coil, BUT it will increase the time to get the coil current to zero. This slows the switching action and is bad for a circuit that wants, or needs, to switch off in a short time. In your post #1, you said you have a FAST pulse, but for me a fast pulse is 10 nanoseconds. For a diesel motor, I reckon a fast pulse would be about 10 microseconds or more. So even with basic equipment you should get good results.
hope this helps.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top