How Distortion Free are the Distortion Measurers? - Page 7 - diyAudio
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 28th July 2012, 02:45 AM   #61
1audio is offline 1audio  United States
diyAudio Member
 
Join Date: Mar 2004
Location: SF Bay Area
Blog Entries: 3
Richiem:
The 510A has been quite stable this week, varying from 10.0000V to 10.0025V max and usually following the ambient temperature. Maybe I'm just lucky with this one.
__________________
Demian Martin
Product Design Services
  Reply With Quote
Old 28th July 2012, 05:21 AM   #62
richiem is offline richiem  United States
diyAudio Member
 
richiem's Avatar
 
Join Date: Sep 2010
Location: Grapeview, WA
In general, the meaningful harmonics will be from 2 up to 10, so a measuring bandwidth of 10X the fundamental is customary. Some industrial, broadcast, telephony, and military systems specify standard low-pass frequencies of 30kHz and 80kHz when measuring in the audio band, and 400Hz high-pass filtering is very helpful in reducing hum and noise in various radio and telephony measurements, not to mention high-end audio gear.
  Reply With Quote
Old 28th July 2012, 03:10 PM   #63
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
Quote:
Originally Posted by Loudthud View Post
I found an old Tektronix 7L5 Spectrum Analyzer to look at the Reading Monitor Output. There was a strong birdie at 48.5kHz when the Monitor was on. That was with a 22kHz filter on the distortion. Monitor on or off didn't make much difference. The options can set it at 22kHz, 30kHz, 80kHz or >500kHz. Those filter selections revealed harmonics of the 48.5kHz and an increase in the residual distortion at low signal levels.
My spectrum analyzer shows a fair amount of junk (spikes of 10 to 20 db above the noise floor) in the 40 to 50kHz region. I attribute these to noise from switchmode power supplies for my computer monitor, computer, battery chargers, and fluorscent lights. I don't know if they are radiated (EMI) or carried on the power lines, or both. The trouble with the hp8903 is that they are counted - summed into the THD + noise calculation. When looking at low distortion devices such as preamps, they swamp the THD result. Now I just use Dick's twin-t to null the fundamental about 50db and then read the fundamentals off the spectrum analyzer and do the THD math manually. This is one of several reasons why most folks say that a THD number alone is not very informative.

Ken
  Reply With Quote
Old 31st July 2012, 04:30 PM   #64
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
Ken, I experience the same thing with my setup.

I blew up a plot with the input of the analyzer hooked to an amp (which was turned off) and started counting harmonics of 60Hz and found that out beyond 3KHz it was all related (as far as I was confident in my interpolation).

If I do a straight loop-back test I don't see the noise.

I've isolated some of the noise to specific sources in my house by turning them off and back on while observing the results. One florescent lamp in the den in particular gives me a lot of noise around 16KHz-23KHz.
  Reply With Quote
Old 31st July 2012, 05:26 PM   #65
diyAudio Member
 
Loudthud's Avatar
 
Join Date: Nov 2007
Location: Near Dallas Texas USA
Florescent light efficiency peaks at around 15kHz so expect to see more of that in the future.

I'm unclear on how to get a percent distortion number from a spectrum analyzer. -40dB would be 1%, -50dB would be about 0.3% but how do you add the two together? Simple addition of root sum of the squares?

Looking at the monitor output of a distortion analyzer, the root is nulled out so there is no reference, and it's auto-scaled so you have no idea of the gain. But you can see a lot of harmonics that otherwise would be burried. The 7L5 only has about 76dB between the Reference level and the noise floor so that's 0.02% or so.
  Reply With Quote
Old 31st July 2012, 05:57 PM   #66
richiem is offline richiem  United States
diyAudio Member
 
richiem's Avatar
 
Join Date: Sep 2010
Location: Grapeview, WA
@loudthud -- the scaling and reference level issues depend on the analyzer. The HP 339 is not a problem, for example, while the 8903 is. But the analyzers have their own issues with noise and self-distortion, so some care is needed in analyzing results at the monitor output.

The dynamic range issues of the 7L5 for example can be dealt with by using an active notch filter to reduce the fundamental without losing track of the reference level for remaining distortion and/or noise products. A good spectrum analyzer is a great thing to have -- wish I had a 7L5 -- just not in the budget right now....
__________________
...................
Dick Moore
  Reply With Quote
Old 31st July 2012, 08:12 PM   #67
TheGimp is offline TheGimp  United States
diyAudio Member
 
TheGimp's Avatar
 
Join Date: Aug 2009
Location: Johnson City, TN
I wonder what kind of problems we will have with LED lights in home use once they are widespread.
  Reply With Quote
Old 31st July 2012, 08:56 PM   #68
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
Quote:
Originally Posted by TheGimp View Post
Ken, I experience the same thing with my setup.

I blew up a plot with the input of the analyzer hooked to an amp (which was turned off) and started counting harmonics of 60Hz and found that out beyond 3KHz it was all related (as far as I was confident in my interpolation).

If I do a straight loop-back test I don't see the noise.

I've isolated some of the noise to specific sources in my house by turning them off and back on while observing the results. One florescent lamp in the den in particular gives me a lot of noise around 16KHz-23KHz.
When I feed the distortion analyzer with a signal from my low distortion signal generator I don't have the spurious noise(s). Only when I feed the amp from the generator then to the analyzer I get the noise(s). I believe a few things are at work here, one I haven't done a good job with emi shielding of the DUT (it's current open air), nor the power supply, and, two, I may have cable impedance mis-match with the analyzer which allows the cables to pick up the noise. Ultimately, the test for the finished amp will be to have a signal that is as clean as the signal generator signal.

Ken
  Reply With Quote
Old 31st July 2012, 09:09 PM   #69
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
Quote:
Originally Posted by Loudthud View Post
I'm unclear on how to get a percent distortion number from a spectrum analyzer. -40dB would be 1%, -50dB would be about 0.3% but how do you add the two together? Simple addition of root sum of the squares?

The 7L5 only has about 76dB between the Reference level and the noise floor so that's 0.02% or so.
Regarding your second statement, my analyzer has 85db of headroom, but, here is the trick: Build Dick's twin-t notch filter to reduce the fundamental frequency by 50 or 60db, then your harmonic fundamentals will show up - the noise floor will drop. My analyzer has a noise floor of approximately -125db so, with a 50 db fundamental null, I can see the full noise floor and any harmonic fundamentals that rise above it. The second part of the deal is to make sure that your DUT is putting out exactly 1vrms or 0db. For a preamp type device, what I do is first hook the DUT up to the analyzer and adjust either the signal generator and or the preamp volume control until the fundamental reads 0db. I turn off the DUT, hook it up to the twin-t, hook the output of the twin-t to the analyzer, turn on the twin-t and then the DUT. Adjust the null of the twin-t to something at or greater than 50, then run the calibration sequence of the analyzer, then run the averages function of the analyzer a stable mean average of 50 and viola you have your spectrum.

Gotta go... how to THD calc later

Ken
  Reply With Quote
Old 1st August 2012, 12:57 AM   #70
klewis is offline klewis  United States
diyAudio Member
 
klewis's Avatar
 
Join Date: Sep 2007
Blog Entries: 36
Attached is my THD calc spread sheet. It assumes the fundamental is set to 0db as described in the prior response. Just change the 2nd through 9th fundamental harmonic values (red) and it will calc the THD.

Ken
Attached Files
File Type: zip THD Calc.zip (14.1 KB, 27 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
FREE Distortion Analyzer imix500 Swap Meet 3 27th July 2010 11:15 PM
Free to good home..HP 302a distortion analyzer-needs work boywonder Swap Meet 8 12th March 2010 01:58 PM
Distortion. bigwill Tubes / Valves 8 14th July 2007 10:47 PM
Non Linear Distortion testing - Harmonic Distortion JMB Multi-Way 0 20th July 2004 02:54 PM
Cable Distortion: Materials Science Free For All Steve Eddy Everything Else 21 24th December 2003 02:01 AM


New To Site? Need Help?

All times are GMT. The time now is 12:48 PM.


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