ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 ... 5 6 [7] 8   Go Down

Author Topic: Cost effective (even for weekend warriors) ways to measure power quality  (Read 19047 times)

Scott Helmke

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1788

 
Caveat, different wall wart-transformers can use different internal transformer construction, so audio frequency response could vary significantly between different lumps.

You might be able to benchmark the wall-wart frequency response driving it with a power amp. This shouldn't hurt the power amp but confirm the amp is happy with the unusual load. You shouldn't have to drive it to full 115VAC to characterize response. 

Planning on doing just that, actually.  Last night I ordered a nice generic 9v wall wart from Jameco, so that I can come up with a good-enough design that doesn't depend on my junk box wall wart of unknown origin.  I don't think I'd need a power amp just to get enough input voltage, but I do want something with a very low output impedance to make the test as accurate as possible.

Back a couple years I was using SMAART with a headphone amp to test various audio isolation transformers. Very handy for looking at distortion at different frequencies and signal levels.
Logged

Scott Helmke

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1788

I am not sure about linearity, but a small wall wart would drop the voltage down to something that won't kill you and is double insulated. Unloaded it should give a fair representation of the sine wave, while I am not sure what you are supposed to do about "harmonics" (what they call dirty power).

Way back at the start... and JR wins... something. Cup of coffee and a muffin, given the budget here?  :)

I tested both my existing 9v wall wart and a brand new Jameco model 100061 wall wart transformer ($8.95) for frequency and phase response, using SMAART and a moderate-sized power amp. Probably not running at a full 120 volts, maybe about half that.  The result? Quite surprisingly flat.

Next step is interfacing to an iPhone mic input. Jameco sells an appropriate TRRS plug, part number 2200954 ($1.95). I made up a two resistor voltage divider - 1Mohm and 1Kohm, for a rather drastic reduction of the 9 volts down to something like 9mV. The iPhone circuitry uses some kind of active load sensing to see if there's a mic attached (sleeve is signal, second ring is ground), and the Internet seems to agree that a resistance of around 1Kohm takes care of that. So, 1K resistor between sleeve and second ring, one side of 9v wall wart output to second ring, and other side of wall wart output through 1M resistor to sleeve. 

The RTA display is an app, "Audio Spectrum Analyzer" by Black Cat Systems ($2.99). Settings I used include Logarithmic display, Hanning window, FFT Length = 131072.

Once I figure out how to fit the resistors into that tiny connector I'll have my pocket-sized $20 power quality meter.

Logged

Scott Helmke

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1788

So that's the hardware and software. I still want to make up a complementary solution to measure neutral to ground.

The rest is interpretation and then solution.

Interpretation is going to be a matter of practice. Measure power in a bunch of locations, take screenshots maybe, and note any issues with power at any of those locations. I've gotten pretty good at reading a SMAART transfer function from a noisy room while tuning speaker systems, so I expect that with practice I'll be able to interpret this stuff as well.

Solution is of course a much bigger issue.  There's not a lot you can do about the power in a lot of situations... though there are maybe some cheats. Maybe only one or two pieces of gear will actually cause problems from noisy/dirty power, such that some filtering would be a reasonable thing to carry around. "Outlet shopping" will be feasible with a portable metering solution, allowing users to find and avoid specific outlets that are noisy.
Logged

Mike Sokol

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3361
  • Lead instructor for the No~Shock~Zone
    • No~Shock~Zone Electrical Safety

So that's the hardware and software. I still want to make up a complementary solution to measure neutral to ground.

Neutral to ground voltages are typically around 1/2 of the voltage drop in the branch circuit. So if you start with 120 volts at the panel, and its dropped to 110 volts at your amp rack while under load, you can have up to 5 volts between the neutral and ground. Of course, this voltage will be modulated by any current draw on that branch circuit. You've also got to watch for swapped Hot-Neutral polarity in the receptacle, since that will cause 120 volts potential between the neutral and ground.
Logged
Mike Sokol
mike@noshockzone.org
www.NoShockZone.org

Scott Helmke

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1788

Neutral to ground voltages are typically around 1/2 of the voltage drop in the branch circuit. So if you start with 120 volts at the panel, and its dropped to 110 volts at your amp rack while under load, you can have up to 5 volts between the neutral and ground. Of course, this voltage will be modulated by any current draw on that branch circuit. You've also got to watch for swapped Hot-Neutral polarity in the receptacle, since that will cause 120 volts potential between the neutral and ground.

Yes, this will actually be the more complex part because the voltage range is more variable. I'm thinking some kind of variable attenuator along with some basic overvolt protection on the output, twiddle the knob until the readings make sense.
Logged

Mike Sokol

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3361
  • Lead instructor for the No~Shock~Zone
    • No~Shock~Zone Electrical Safety

Yes, this will actually be the more complex part because the voltage range is more variable. I'm thinking some kind of variable attenuator along with some basic overvolt protection on the output, twiddle the knob until the readings make sense.
I would suggest series limiting resistors with clamping diodes
Logged
Mike Sokol
mike@noshockzone.org
www.NoShockZone.org

Scott Helmke

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1788

I would suggest series limiting resistors with clamping diodes

That would keep it below about half a volt, though I've been toying with the idea of using clamping LEDs before attenuating the signal the rest of the way down to iphone mic level. That way you know there's a problem right from the start.
Logged

Mike Sokol

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3361
  • Lead instructor for the No~Shock~Zone
    • No~Shock~Zone Electrical Safety

That would keep it below about half a volt, though I've been toying with the idea of using clamping LEDs before attenuating the signal the rest of the way down to iphone mic level. That way you know there's a problem right from the start.

Also remember to calculate when happens if you accidentally plug into a 120-volt receptacle mis-wired with 240-volts. I blew up a nice Sennheiser RF receiver one time by plugging into a "special" outlet they wired for the janitor's floor polisher in the gym. Yup, this was a brand new NEMA 5-15 receptacle in the wall wired up with 240 volts instead of 120 volts. I've seen this several times at gas stations, contractors compressors, and some of the PA systems I run for a British sound company in the area.   
Logged
Mike Sokol
mike@noshockzone.org
www.NoShockZone.org

Scott Helmke

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1788

Also remember to calculate when happens if you accidentally plug into a 120-volt receptacle mis-wired with 240-volts. I blew up a nice Sennheiser RF receiver one time by plugging into a "special" outlet they wired for the janitor's floor polisher in the gym. Yup, this was a brand new NEMA 5-15 receptacle in the wall wired up with 240 volts instead of 120 volts. I've seen this several times at gas stations, contractors compressors, and some of the PA systems I run for a British sound company in the area.

That's where I'm hoping JR's awesome outlet tester finds a manufacturer.   8)
Logged

John Roberts {JR}

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 16831
  • Hickory, Mississippi, USA
    • Resotune

That's where I'm hoping JR's awesome outlet tester finds a manufacturer.   8)
I dropped the 240V feature from my final cut of PCB because it took up too much room. Right now the OD-1 PCB fits neatly inside a standard off the shelf plug.

Just for chuckles I tested my 240v detector circuit design anyhow, and shared it with the most recent company to gong me..

The picture with the neon bulb glowing is across 240V, the second picture across only 120V is dark. Circuit is dirt simple, only two resistors forming a simple divider such that 120V divided is below neon threshold and 240 divided isn't.

The neon lamp is too thick so doesn't fit inside between the plug housing and PCB, and 200+V SMD resistors need to be 2012. I could almost cut a hole in the PCB for some extra clearance for the bulb thickness but the larger resistors to handle high voltage take up all my PCB real estate already.

JR

PS: I have my final cut PCB and parts waiting on my bench but I am forcing myself to do my taxes first...  and I hate doing taxes so this could take some time with me procrastinating. 
Logged
Cancel the "cancel culture". Do not participate in mob hatred.

ProSoundWeb Community


Pages: 1 ... 5 6 [7] 8   Go Up
 



Site Hosted By Ashdown Technologies, Inc.

Page created in 0.108 seconds with 21 queries.