Go Back   Home > Forums > Design & Build > Equipment & Tools

Equipment & Tools From test equipment to hand tools

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 13th May 2012, 12:44 PM   #21
diyAudio Member
 
Join Date: Sep 2004
Location: UK
They would need to be trimmed using 0-10pF trimmers with a square wave test input.
  Reply With Quote
Old 13th May 2012, 01:36 PM   #22
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
Quote:
Originally Posted by PChi
You have used rather high value resistors.
Yes. Try changing the 100k resistors around the opamps for 10k, or even less. That might be enough to get the bandwidth you need.
  Reply With Quote
Old 13th May 2012, 04:07 PM   #23
akis is offline akis  United Kingdom
diyAudio Member
 
Join Date: Jul 2008
Hi thanks for the excellent links (still reading).

The problem was (still is) the resistor network I have used to divide the input signal by 10 or by 100. The resistors were closely placed together and it seems at higher frequencies the signal simply jumped from one to the next thus not being attenuated as it should. I replaced the resistors placing them wider apart and this made a big difference, but still is not perfect.
  Reply With Quote
Old 13th May 2012, 04:17 PM   #24
jitter is offline jitter  Netherlands
diyAudio Member
 
jitter's Avatar
 
Join Date: Aug 2004
Location: Eindhoven, The Netherlands
Quote:
Originally Posted by akis View Post
So I think that when I meantioned "floating voltages" some people instantly thought "many hundreds or thousands of volts in high tension photovoltaic circuits, or tube amplifiers" but I say again this is not the reason to try and do this.
Well, I mentioned the high voltage part for several reasons:
- it's one of the reasons to use diff probes;
- other people are going to read this thread too (now or in a couple of years);
- you may want to measure higher voltages at some point in the future.

BTW, 100 Vp can hardly be called low voltage (= 200 Vpp or 141 Vrms).
  Reply With Quote
Old 13th May 2012, 04:23 PM   #25
jitter is offline jitter  Netherlands
diyAudio Member
 
jitter's Avatar
 
Join Date: Aug 2004
Location: Eindhoven, The Netherlands
Quote:
Originally Posted by akis View Post
Hi thanks for the excellent links (still reading).

The problem was (still is) the resistor network I have used to divide the input signal by 10 or by 100. The resistors were closely placed together and it seems at higher frequencies the signal simply jumped from one to the next thus not being attenuated as it should. I replaced the resistors placing them wider apart and this made a big difference, but still is not perfect.
Google "diy differential probe". I did and I found some threads in another forum from which I get the impression that getting a flat frequency response of a diff probe is very difficult. Hence the high price. Perhaps you can find some tips'n'tricks there too.
  Reply With Quote
Old 13th May 2012, 04:26 PM   #26
diyAudio Member
 
Join Date: Sep 2004
Location: UK
"The resistors were closely placed together and it seems at higher frequencies the signal simply jumped from one to the next thus not being attenuated as it should."

Is this in a "breadboard" layout, with lots of stray wires? In any case you are suffering form feedback / oscillation which you will only resolve by looking at it with your 'scope!

This kind of gain / precision will only work with a pretty tight, clean layout and a ground plane. A prototype can get close if you use the "dead bug" method on some bare pcb board, grounded.

See around page 4
http://www.dartmouth.edu/~sullivan/prototyping.pdf
  Reply With Quote
Old 13th May 2012, 04:35 PM   #27
DF96 is offline DF96  England
diyAudio Member
 
Join Date: May 2007
You can add capacitors to the voltage divider but don't overdo it as it will cause HF loading. This is a standard technique, but must be done carefully.
  Reply With Quote
Old 13th May 2012, 06:26 PM   #28
akis is offline akis  United Kingdom
diyAudio Member
 
Join Date: Jul 2008
I take it all back. It is not the resistors proximity on my divider network or the breadboard itself.

I just did a cold test, and was amazed at the results. I created a simple resistive divider with two resistors of 820K each, as shown in the attached picture. For 200Hz I read +/- 1V on the scope. At 200KHz I read +/- 0.8V and at 500KHz I read +/- 0.5V. Replacing the resistors on the divider with 10K values, and talking them completely out of the breadboard makes no difference.

I know the probe has a small capacitance but to be unable to read a 100KHz frequency from a simple resistive divider of 20K is news to me. I cannot simulate this on my simulator, any value of parallel capacitance that I add works only for a single value of resistors on the potential divider, whereas the scope produces consistently incorrect results regardless of the values of the resistors.


Edit: the capacitor and 1M resistor were not on the test circuit, I have added them to the simulation to try and recreate the phenomenon.
Attached Images
File Type: jpg differential-probes.JPG (19.4 KB, 97 views)

Last edited by akis; 13th May 2012 at 06:31 PM.
  Reply With Quote
Old 13th May 2012, 07:01 PM   #29
diyAudio Member
 
Join Date: Sep 2004
Location: UK
AKIS, you really must slow down and be more methodical.

1) always check things at DC first. That eliminates C, L and checks that resistor ratios are as you expect.

2) 100pF at 100Khz is Z=16K! Completely KILLING your divider. Do your sums first!

For a 1% error, any loading must be >100 times the source. So with 20K resistors you will need in the order of >2M load

3) Do it at DC FIRST! (Did I already say that?) :-)

Last edited by cliffforrest; 13th May 2012 at 07:07 PM.
  Reply With Quote
Old 13th May 2012, 07:34 PM   #30
akis is offline akis  United Kingdom
diyAudio Member
 
Join Date: Jul 2008
The picture above is a simulation trying to recreate the error that I got by testing - I added the cap and resistor to reproduce the error in simulation.

There must be something wrong with my oscilloscope/probes or I am missing something. My scope is rated 20MHz.

Here is another picture of the *real* test I did. If you have an oscilloscope handy can you please tell me what it reads? It should read +/- 1 V. But I read 0.5V at 500KHz and 0.8V at 200KHz.
Attached Images
File Type: jpg differential-probes.JPG (12.3 KB, 91 views)
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Measuring ultra high voltages with DMM Alexontherocks Equipment & Tools 9 14th April 2012 02:16 PM
measuring voltages and currents simple se pforeman Tubes / Valves 1 10th August 2009 02:38 AM
cheap usb oscilloscope for measuring amplifiers umut1001 Class D 27 5th September 2007 01:33 PM
Floating voltages eeka chu Solid State 12 29th January 2004 03:57 PM
Floating voltages eeka chu Parts 2 27th January 2004 09:10 AM


New To Site? Need Help?

All times are GMT. The time now is 11:47 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2