Measurement Woes

[augerpro]

You do have the stereo and computer plugged into the same outlet now correct? Now unplug EVERYTHING else from the stereo and do your measurements. If you have anything plugged into a different outlet and directly or indirectly hooked to the stereo you'll still have a ground loop. After you take the measurements you can run around trying to find teh problem, my money is on the coax not being grounded well outside.

[cjd]

Amp stage probably will need ground lifted - the switching PS in the computer can do odd things sometimes if the amp is also grounded.

C

[augerpro]

Chris are you saying use a 2 prong adaptor on the amp?

[ThomasW]

Chris are you saying use a 2 prong adaptor on the amp?

Yes...

[Blazin]

Chris are you saying use a 2 prong adaptor on the amp?

uhm.. a two prong adapter? as in from 2 prongs to 3..? It is a 2 prong plug already..


The amp and computer are plugged into the same power strip.. and I plugged the Mic pre-amp into the outlet on amp, I could move this to the power strip, but isn't this the same theoreticly?

I had the following:

Amplifier Input:

3 Optical - can't cause ground loop
1 RCA - Computer in

Amplifier 'A' Speakers all connected
Amplifier Subwoofer Pre-out connected
Amplifier 'B' Speaker front, powers DUT


Now that I type this.. I just realized the Powered Subwoofer is plugged into a different outlet form the rest of the equipment.. I'll check this later..



I'm seeing the final suggestion here is:

Move amplifier, and computer into an isolated environment that only involves the bare essentials to clear up this ground loop..

I will begin doing this as soon as possible.. with much reluctancy as I was hoping to be able to make these measurements without dismantling my HT Setup..

Oh well.. its a plan of isolating the components or spending more and more time on this nonsense..

[cjd]

THis is why the "unplug everything else from the amp" step is crucial.

Subs LOVE to add ground loop... and you definitely don't need it in the mix for your measuring.

C

[Dennis H]

Maybe try a cheater plug on the computer. You've got a ground loop somewhere assuming the amp doesn't hum with nothing connected (it doesn't right?).

[Blazin]

Maybe try a cheater plug on the computer. You've got a ground loop somewhere assuming the amp doesn't hum with nothing connected (it doesn't right?).

Yea i can only hear the hum when the computer is plugged into the Amp.. when the Coax/HDMI is plugged into the Cable box..

It HAS to be the Sub and/or the Digital Coax cable plugged into the amp.. the other connections are optical..


And if that doesnt work.. I'll try a 3-2 prong adapter for the computer..

[augerpro]

What's the digital coax for? Unplug it. And what do you mean about the HDMI/coax for the cable box? I thought you unplugged all that already?

[Blazin]

So last night I unplugged everything from the amplifier except the 'A' speakers..

Still there..


Now does this make sense:

Plug into either Line in or Line out with the minijack to RCA cable and immediately.. 60Hz + harmonics appear on SynRTA.

Plug them both in and then loop back, 60Hz + harmonics gone.

SynRTA has an output level volume, if I do loopback and I raise the output volume to max, I achieve the most flat loopback FR at around 0 dB, lower output levels introduce a less than ideal FR (not any one peaking point, but lower Amplitude and much more variance)


I have lots of time now to debug this as the semester comes to a close and I have already handed in a Final Design Review utilizing Manu. Specs to model some crossovers..


Next step:

Move amp and computer to garage, isolated from all other electronics.. I'm now thinking the Line/Line out of this sound card is causing the problems, its not the best one in the world, its a bit quirky.. perhaps this is a known issue..

I'll replace it for the time being with my good old SB Live! 5.1.. Hardware cards over Software any day!

[Blazin]

Alright so hear's a fun step in this process:

Farfield Measurements appear to be exactly as expected..(except for my MIDs but thats because the expected is for 1 driver and I have 2 in parallel)

I performed Farfield Gated Measurements on axis with the tweeter same gate and mic position for all drivers.. The high frequencies appear well, while the low frequency has very little resolution -thats fine and understood-.

From my reading, the nearfield measurements should have a higher amplitude than the farfield.. Well.. to start my farfield amplitude peaks are about 60-90dB(i'd have to actually inspect a little) where as my terrible nearfield peaks at maybe about 3dB...

Yea I can't figure that one out.

2.) I have figured out the problem with my amplifier.. can't explain it.. but figured something out.

1- Speaker set B, if turned on causes 60Hz harmonics (so the whole time i had been messing with everything, that was the cause.)
2- Any other listening setting other than something with DSP causes 60Hz.

Conclusion:

Well I don't know enough about the inner workings; however, I know that Speaker B input cannot perform any DSP only DIRECT Stereo.. it appears that the DSP inside the amplifier, rectifies any harmonics.

The problem.. haha well thanks to the good old folks at Onkyo, I have the ability to set a crossover point anywhere from 200-40Hz.. cool, except that means my SynRTA Freq Response curve has a roll off at 40Hz.. (SUPER! -sarcasm-) I can't turn crossover off.. at least I couldn't figure out how(I tried saying NO to Subwoofer, thinking it'd let me select 0 crossover, but I guess Onkyo assumes no speakers will support frequencies lower than 40Hz.. stupid.)



Oh annnd to top this all off.. my nearfield.. is still reading 60Hz peak on all speakers, therefore it is causing my mids and sub measurements to roll off muuuch higher than predicted..


Please someone tell me what the hell is going on.. and is there anyone in MA/NH with a measurement setup that can help me? I really can't afford to waste anymore time on this craziness.

[Dave Bullet]

Blazin,

It might be best if you give a detailed rundown to us on "what is connected to what" with your now "minimalist" setup. All you should have is your amplifier and speakers under test connected to your soundcard (removing all other equipment, subs, cable etc...). Sometimes going over this basic recheck will help you identify anything erroneous.

What I would do is just with the speaker under test connected to the speaker A output (and no soundcard connected), select an analogue source (ie. CD) and crank the volume to see if you can see and hear a 60Hz signal being played. You want to try your test condition in the previous post - setting the amp to the "creates hum" config and remove / connect the soundcard. If without it connected, you know it is within the amplifier. Might be easier then to borrow another test amp. IT seems odd that the hum only occurs when a signal is being played through the soundcard.

My experience with ground loop hum (assuming this is it) is once a connection is made - you hear it (it is a residual current leaking across the line inputs between devices - it doesn't need any other stimulus such as a source or signal being played to create it).

Dave.

[Blazin]

Blazin,

It might be best if you give a detailed rundown to us on "what is connected to what" with your now "minimalist" setup. All you should have is your amplifier and speakers under test connected to your soundcard (removing all other equipment, subs, cable etc...). Sometimes going over this basic recheck will help you identify anything erroneous.

What I would do is just with the speaker under test connected to the speaker A output (and no soundcard connected), select an analogue source (ie. CD) and crank the volume to see if you can see and hear a 60Hz signal being played. You want to try your test condition in the previous post - setting the amp to the "creates hum" config and remove / connect the soundcard. If without it connected, you know it is within the amplifier. Might be easier then to borrow another test amp. IT seems odd that the hum only occurs when a signal is being played through the soundcard.

My experience with ground loop hum (assuming this is it) is once a connection is made - you hear it (it is a residual current leaking across the line inputs between devices - it doesn't need any other stimulus such as a source or signal being played to create it).

Dave.

Alright.. I have pretty much done this already.. I'll try to make a test matrix and post it here so I don't get the same questions over and over and over and over again.


Minimal setup means minimal, Amplifier with 1 connection, the DUT. Computer with 3 connections, Line IN, Optical OUT, VGA OUT...

Amplifier has nothign connected to it..

I gotta do this again and make sure I test every little tiny thing and try to make these measurements.. there could be somethign that I am overlooking...

[Blazin]

Alright..


Someone help me.

this 60Hz thing.. It clearly cannot be ground loop.. I think it's the measurement microphone/preamp..

Why?

Well since I'm seeing the same garbage at the same frequencies I tend to assume it has something to do with a specific element..

Sure I can get ground loop to happen..

BUT.... using an Optical cable should eliminate this.


It has not.. hence microphone/preamp?


It doesn't make sense to me that a strong peak at 60Hz would cause cancellation at frequencies near 60Hz.. in my mind 60 Hz peak should then add to whatever frequency response is being recorded.. seeing a peak at 60Hz while the frequencies (which should be resonating ) at 30-50 and 70-200 Hz should remain "flat" albeit lower than the peak, some where around the level of the higher frequencies.

This is not the case.

Why am I always getting cancellation at 70-200 Hz and 5-50 Hz as if the 60Hz Peak skirting is cancelling out the frequencies at those locations pulling them down to FAR(-) amplitudes.. that doesn't make sense.

Someone please explain to me if that thinkin is incorrect, I discussed it with my signals and systems professor, he agrees.. in no way shape or form should a ground loop be causing signal cancellation.. it should be adding a peak at 60Hz but not casuing cancellation.

[Hdale85]

Might want to edit your post for language.... Also have you gotten the mic calibrated?

[ThomasW]

In situations like this the only option is to start replacing components until the problem goes away.

[Blazin]

Ugh.. my fear is becoming a reality.. i've been trying to measure frequency response for 3 months.. this is so terrible.

[---k---]

I would take your gear into the lab at school and see what happens. Maybe mix in some different gear there, depending on what they have. Maybe breaking everything down and a fresh start will help.

[Blazin]

I would take your gear into the lab at school and see what happens. Maybe mix in some different gear there, depending on what they have. Maybe breaking everything down and a fresh start will help.

Yep, moved everything into the lab today..

Problems:

I have to use my amp.
I have to use the preamp/microphone I sourced for the school(They didn't have audio measurement equipment)
Only new piece of gear: Computer.. computer has nothing to do with this problem.

Wallin's jig has nothing to do with it.. since I get the same results with or without it(ie Claudio cable setup)

Amplifier or Mic or Preamp.. great.. how do I test that.. swap equipment.. yeah kind of in a bind to do that as I don't have a way in which to do that.

[ThomasW]

It's unlikely that it's the mic, they either work fine or they're don't work at all.

I read where one person had a defective MP13. When he checked the output voltages with a DMM, it wasn't putting out the necessary amount of phantom power needed to drive the mic.

[Dave Bullet]

I can understand your frustration! But I'm sure we can solve it.

Does your amp have pre-out / mains in jumpers?

If so...
Mic -> pre-amp
Pre-amp -> amplifier mains in jacks

Diconnect everything else (including computer).

Crank the volume on the power amp (slowly). If you can't hear any hum then you are right in that it can't be a ground loop caused by the mic/pre-amp combo and must be another anomaly.

Dave.

[Dennis H]

Sure I can get ground loop to happen..

BUT.... using an Optical cable should eliminate this.

Not necessarily. You still have to get the signal back into the computer with plain ol' wire so a ground loop with the computer is still possible. I'm not saying that's the cause (I have no idea what is) but it's a possiblity.

[Blazin]

So here's a fun one.

My professor and I have been trying to debug this for an hour and a half..

Amplifier, tested with a non-inductive 8ohm resistor.. measured the output.. it appears flat as low as 30Hz, no change in gain. All using a signal generator.

Check.


Lets try this with SW sine generator:

FAIL!

Amplitude is lower with lower frequency.. so the output of the sound card is attenuating the output as the frequency decreases.. the 60 Hz peak.. unexplained.. I tried generating a 60Hz sine wave.. it had lower amplitude than 100Hz..


Hah..

Hookup unbalanced output of MP13 to oscilloscope.. hah.. 60Hz..

So the preamp is causing the 60Hz peak.. and the sound card is causing the Rolloff after about 300Hz..

Preamp's power supply reads at 18.76 V... preamp needs 12V DC.. stupid PS

Bench PS with Barrel plug at 12V to fix that.


HOLY GOD IT'S WORKING

SETUP:

Line OUT L Channel -> Split -> Amplifier IN & Line IN LEFT
Line IN R Channel < - Microphone Pre-amp L Channel OUT


Works


SO here was the problem:



And how I fixed it.

[stangbat]

I've been following this wondering if you'd get it figured out. So if I'm understanding correctly, you didn't need to change any components other than the PS to the mic preamp? And the main hookup change was the split of the left SC line out to the amp and the left SC line in?

[Blazin]

I've been following this wondering if you'd get it figured out. So if I'm understanding correctly, you didn't need to change any components other than the PS to the mic preamp? And the main hookup change was the split of the left SC line out to the amp and the left SC line in?

Preamp PS was borked, so that was replaced, and yes..

SC Line out Left Channel went to both Amplifier Line In Left and SC Line In Left..

The SC Left channel out was the reference.. I'm thinking wallin's jig was supposed to use the amplifier's output signal as the reference.. well this wasn't working.. so I used the SC output as the reference.. much better.. as it actually WORKS

Hope this helps if you're having a problem