Am I using my oscilloscope correctly? / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Am I using my oscilloscope correctly?

1 2 next

Move Thread

LAN_403

hak8or

Sat Feb 13 2010, 07:15AM

Registered Member #2025 Joined: Fri Mar 13 2009, 05:39PM
Location:
Posts: 16

I am trying to see the waveform for a IRFP450n being switched at about 17khz with VDS at 12v from a computer power supply. I put a 3.9 ohm resistor from the mosfet drain and the 12v+ from the power supply.
The mosfet gate is driven at 5 volts from a pic.
The probes gnd is to circuit gnd, and the probe is to the pin of the drain on the mosfet.

The problem is that the scope shows me this
http://picasaweb.google.com/lh/photo/YJBnosJBopTOb4deF_z_Wg?authkey=Gv1sRgCOq2xLrfqJ_eAg&feat=directlink.
The top line is the +5v from the pic pin, and bottom is the drain.

Upon closer view it seems that it is resonating?

http://picasaweb.google.com/lh/photo/26icZvAL5yzxG6jK7CsGyQ?authkey=Gv1sRgCOq2xLrfqJ_eAg&feat=directlink

When I put in a speaker between the mosfet drain and the resistor, I get the high pitched squeel as expected which is this
http://picasaweb.google.com/lh/photo/nUs43958QhszqgjViJ4Dag?authkey=Gv1sRgCOq2xLrfqJ_eAg&feat=directlink

I am trying to capture the waveform that the mosfet produces, and am expecting the wave to go from 0v to 12v back to 0v, but instead it seems to resonate at around 11.5 volts. Can anyone please clarify? Maybe I am putting the probes in the wrong spot? Thanks

Mattski

Sat Feb 13 2010, 08:01AM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

What is the DC level of the voltage at the drain node? When the FET is turned off, it should be at 12V, then when it's turned on it should go to 0V.

You might also try with a higher gate voltage than 5V, it should be enough but I'd try 12V. You can test it without doing a square wave driver, just put a wire on the gate, touch the wire to 0V, then 12V, and check the DC voltage at the drain in both cases.

Steve Conner

Sat Feb 13 2010, 09:39AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Hey hak0r

Looks like something wrong with your scope or your probe. Maybe the probe is broken, or you forgot to turn off the Ground button on channel 2.

Try swapping the two probes around on your circuit, so you're looking at the drain waveform on the top channel, and the gate waveform on the bottom one. Also try probing the scope's calibrator with both channels. This should help you track down the problem.

hak8or

Sat Feb 13 2010, 03:12PM

Registered Member #2025 Joined: Fri Mar 13 2009, 05:39PM
Location:
Posts: 16

Thank you for the replies.
The voltage at the drain node is around 11.5 volts.

Why would I turn off the Ground button on channel two?
I will try a higher gate voltage, change the probes around,and try the calibrator.

radiotech

Sat Feb 13 2010, 03:51PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

What I see is ringing in the drain to ground waveform. What isn't shown is your wiring arrangement, lead lengths in gnd return to power supply.
The one with the speaker is exactly how a speaker alone at 17 kHz
would act if driven by a fast rise waveform.

About the sqealing noice you heard. Last time I could hear that well
Cuba had missles!

hak8or

Sun Feb 14 2010, 03:48AM

Registered Member #2025 Joined: Fri Mar 13 2009, 05:39PM
Location:
Posts: 16

My first try at eagle, excuse any mistakes I seem to have made

I put the probe to see the waveform that drives the mosfet between the mosfet gate and the 2n2222.
I get the proper looking waveform as seen in my first post, the pulses at 5v.

I put a speaker instead of R2 and it gave the high pitched squeal as I thought it would. (Gave me a chuckle radiotech :P )

I then tried the see the waveform of the mosfet turning on, and was unable to do so. I put the ground of the probe to the circuit ground along with trying the probe at other points in the circuit.
I tried between R2 and +12, which gave me as I thought, a 12v line.
Then R2 and Q1 but I did not get any waveform, and the scope did not trigger.
Then between Q1 and gnd, which as suspected also gave me no trigger.

I thought that putting the probe between R2 and Q1 would give me a waveform? I know the mosfet is turning on because I hear the high pitched squealing when I replaced R2 with a speaker. Also, R2 gets slightly warm after a few minutes so I presume it is running with the resistors.

Picture of oscilloscope settings
I think they are all set correctly. I did the probe adjust and both channels are working fine.

So, am I hooking up the probes wrong?
Am I setting something wrong on my scope?
Is there a general way set up the probes on a scope?

I apologize if these questions are too simple or insulting to ask :P
Thanks again for any help

Mattski

Sun Feb 14 2010, 06:53PM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

Putting the leads across the drain resistor is supposed to give you a waveform, the voltage drop is the current through the resistor. If you don't have a changing voltage there it means you don't have a changing current in your MOSFET. Is that what you mean by "between R2 and Q1"?

One thing you need to check though is if your circuit's power supply is isolated from house ground, because your oscilloscope's ground probe is usually directly connected to that. Which means that if your circuit is not isolated you will be short circuiting one of your circuit nodes, possibly blowing a fuse if there is no current limiting.

Is Q1's gate signal going between 0V and 12V?

Have you tried swapping Q1 for a new FET?

You almost never want Ground coupling on your scope because you will not see any signal this way. DC coupling is what you should usually start out with. AC coupling removes any DC bias from the signal, you generally only want to use that once you have verified the DC level of a circuit.

Steve Conner

Sun Feb 14 2010, 06:57PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Oh, remember that the ground clips of the two scope probes are connected to each other, and also to the oscilloscope's casing and house ground. The scope is not like a DMM where you can clip the two leads to anything you want: You can't clip the scope's ground clips to any voltage other than ground, or weird things will happen.

Downright dangerous things too, when you work on higher powered equipment.

Some DSOs have independently floating channels, but that's the exception rather than the rule. I have a Tektronix 222A like this, and though it's ancient and slow, it's still very useful because it can make some measurements that an ordinary scope simply can't. You can look at the tiny voltage across an emitter resistor in a power amp on one channel, and the output voltage of the power amp on the other!

hak8or

Sun Feb 14 2010, 07:50PM

Registered Member #2025 Joined: Fri Mar 13 2009, 05:39PM
Location:
Posts: 16

What I meant by between R2 and Q1 is that the probe is attached to the wire between Q1 and R2. The ground wire from the probe is allways connected to ground.

For all my attempts, I have been putting both the probes ground alligator clips to ground, as if the GND wire.

There is a 12V gate signal going from the 2n222 to the mosfet, as seen with the scope, but it seems that it does not drop all the way to 0v, and instead to around 2v or so. I will try wiring it all from scratch again on my protoboard.

What do you mean by the leads across the drain resistor? Conner said to not put the gnd lead from a probe to anything but ground, and I tried putting the probe to the connection between R2 and Q1.

I will try another fet now.

Thanks for the info Conner, did not know that the probes ground were connected to each other and to the mains ground. :D

hak8or

Fri Feb 19 2010, 05:55AM

Registered Member #2025 Joined: Fri Mar 13 2009, 05:39PM
Location:
Posts: 16

SUCCESS ! :D

Ah, after a few days of being away I tried a Logic level fet, the IRLI530G.
I checked the gate drive from just the mcu, looked good, ov to 5v square wave.

Then I put in a resistor from the Drain to +12v, and checked gate again. I got a tiny slope, so I presume that that is the gate capacitance I am seeing. I stick the probe to the drain, between the resistor and Drain, and surpise suprise, I get a perfect 12v square wave!

I was so amazed and filled with joy that I almost slipped off my chair! :P

I changed the frequency of the PIC to 17KHZ, tried again, and yes, still a good waveform!

I plug in a flyback, and what do you know? I got that nice little breeze and smell. I looked at my scope, the wave going everywhere, high frequency since waves. I need to put in a diode to prevent the distruction of my fet :P

The little arcs were amazing, so much satisfaction! :D
I apalogize for all the horrendous spelling I used, I am too filled with joy right now to look back :P

But I do have a question that still remains.
When I had the transistor supplting 12v to the earlier 450 mosfet, why did the mosfet output generate a very tiny square wave, going to 11.5V high, and 10V or so low?

I still don't understand why it wouldn't go back to zero, as the gate waveform seemed fine.

Thanks anways! :D
Thanks for the help, you all provided me a great deal of assistence, and even encouraged me enough to take a shot at eagle.

Thanks again :D

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob