Gdt output voltage / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Gdt output voltage

previous 1 2

Move Thread

LAN_403

Dr. Slack

Thu May 07 2015, 03:04PM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

You can check your scope accuracy by putting in a known voltage. Measure a (say) 12v battery with your meter, then see how far the trace moves on the scope when you connect it, use DC coupling for this of course. I often do this just for confidence (just because I'm paranoid doesn't mean that things always work), or for checking the gain of a new probe.

Unlike a DMM where you would expect the bandwidth be falling over at GDT frequencies, even the meanest 'scope should be reading the same at that frequency as it does at DC.

Your waveform looks quite nice, sharp transitions and no overshoot.

Scope Gotchas

Be aware that most scopes have an adjustable Y gain, it sounds like you've found this already. Did you know there is a 'cal' position which is usually fully anti-clockwise (or is it clockwise, it should be marked, and mine has a detent there as well) on the adjust knob, for which the volts/div on the range switch is correct. Many a time I've got dud results by not noticing that I'd left my input gain 'uncal'.

Be aware that most scope *probes* have a flatness adjustment, which is significant at GDT frequency. Adjust your probe for flatness with a 1kHz square wave (many scopes have a cal output for this). It must be 1kHz, with GDT frequencies you won't see the flatness error, it will just trash your gain accuracy if mis-adjusted. If you are not using a probe, then you may think you're immune from this error. However the reason we use probes is to reduce the scope's (and the input cable's) effective input capacitance. If your source is too high an impedance, the input capacitance may load down the high frequencies.

In case you're wondering, no, you can't read a scope trace to get a measurement of a few percent accuracy. That's not what it's for.

furnace

Thu May 07 2015, 03:15PM

Registered Member #4992 Joined: Wed May 23 2012, 03:57PM
Location:
Posts: 108

Thanks grant, but these divisions I can usually adjust however I want to, so how would I know that it is in the right position and certainly there must be a formula?

GrantX

Sat May 09 2015, 07:44AM

Registered Member #4074 Joined: Mon Aug 29 2011, 06:58AM
Location: Australia
Posts: 335

furnace wrote ...

Thanks grant, but these divisions I can usually adjust however I want to, so how would I know that it is in the right position and certainly there must be a formula?

Yeah, you should be able to adjust the volts per division setting, and as you do the signal on the screen will get taller or shorter. If you're measuring a 5V square wave, it should take up 5 vertical grid squares at 1 volt per division, and 2.5 at 2 volt per division. A 20V gate drive signal will be 4 grid squares tall at 5 volt per division and 2 squares tall at 10 volt per division. So you don't really need to remember any formulas, I think of it more like a map: the grid squares always have a certain scale, like 1 kilometre per square or whatever. The time divisions are pretty much the same, you'll be looking at a certain number of milli/micro/nano seconds per division.

As Dr Slack said, there are a few "pitfalls" with analog scopes. You'll have to check that the adjustable Y position is on the calibrated setting (marked with an indent or 'cal' mark), and checking the accuracy of the scope against your DMM with a known voltage source like a battery.

Honestly, I'm struggling to remember this stuff, its been about 4 or 5 years since I bought my DSO (which simply displays the peak-to-peak, RMS and DC offset voltages, as well as duty cycle and frequency in plain numerical values. No need for divining the truth from soggy tea leaves or crystal balls).

I hope that helps you out (and hopefully I didn't botch that explanation too badly), if you get stuck it might be a good idea to post a photo of your scope's control panel, then people will be able to see what setting you're using and give further advice.

Dr. Slack

Sat May 09 2015, 08:53AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

GrantX wrote ...

Honestly, I'm struggling to remember this stuff, its been about 4 or 5 years since I bought my DSO (which simply displays the peak-to-peak, RMS and DC offset voltages, as well as duty cycle and frequency in plain numerical values. No need for divining the truth from soggy tea leaves or crystal balls).

I don't use the magic numerical outputs of DSOs unless I have a really simple waveform like a pure sine or square. A 'real' waveform is usually far more complicated. Taking a reading on an analogue scope means estimating the divisions to get 5% accuracy if you're lucky. Taking the numbers from a DSO means getting 3 or 4 significant figures, but then wondering if it's measuring exactly the part of the waveform you'd like it to be measuring. Peak to peak is often not waht's wanted if there's slope or minor overshoots. I suppose setting the cursors to the waveform, and then reading out the cursor positions gives you more certainty about what's being measured, but it still gives you the erroneous pseudo accuracy. Now RMS, if it's relevant to the waveform you are seeing, that is handy to have.

previous 1 2

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