Matching levels on 2ch amp with separate gain for each channel

[Room_101]

I have a 2 channel amp with separate gain knobs for each channel. I've been using it with the gain levels at max, but I'd like to lower them so I can get more usable range and finer increments from the stepped volume control on my preamp. How would you suggest setting these gain knobs so that both channels are equal? Some possibilities I have considered:

1) Visually set them to the same position (e.g. both at 2 o'clock). Is this precise enough?

2) Play a sine wave test tone (e.g. 500hz or 1khz), pink noise, or white noise, and tweak the knobs until the sound seems to be coming from dead center between the two speakers. (I guess this would partially compensate for acoustic asymmetry in the room, but only at the frequency of the test tone if using a sine wave.) Again, would this method be sufficiently precise?

3) Use a multimeter to verify exactly equal output while feeding the amp a test tone, white noise, or pink noise.

4) Use an SPL meter and play a test tone, pink noise, or white noise through one channel at a time - tweak until the measurement at the listening position is the same from each channel. (Again, presumably this would partially compensate for acoustic asymmetry).

5) Similar to number 4, but use a calibrated measurement microphone and REW instead of an SPL meter.

EDIT: 6) Since this is a McIntosh amp, I could also play a test tone, white noise, or pink noise, and tweak until both channels read the same on the big blue VU meters. Are these meters sufficiently precise to use them for this purpose?

Is there another, better method I haven't thought of? Which method would you recommend?

Also, if #4 or #5 is the optimal method, which alternative would you suggest as the next best option for now, since I don't have an SPL meter or measurement microphone yet (I do have a multimeter)?

Thanks in advance for any and all help! The generous sharing of knowledge here is much appreciated!!!

 

[Sokel]

3 to begin with.
If your multimeter is not a true RMS one check it's specs for accuracy,or measure to 400-500Hz and no higher.

 

[fpitas]

Honestly, I'd try 1) first. It's not the most scientific, but the fact is that your ears probably don't have precisely the same response. So, you tweak until a mono signal sounds centered.

 

[Room_101]

3 to begin with.
If your multimeter is not a true RMS one check it's specs for accuracy,or measure to 400-500Hz and no higher.

IIRC, it's a pretty nice Fluke meter, so I think it's likely to be true RMS, but I will check the model number when I get home.

For this method, would I just disconnect the speakers, play a test tone/noise, and measure AC voltage at the speaker terminals on the amp? I thought I read somewhere that it might be a bad idea to send a signal to the amp without speakers connected, but I can't remember why.

Should I use sine wave tones, pink noise, or white noise? If sine wave tones, should I be check at various frequencies or is one enough (e.g. @ 400Hz or 1kHz)?

 

[Room_101]

Honestly, I'd try 1) first. It's not the most scientific, but the fact is that your ears probably don't have precisely the same response. So, you tweak until a mono signal sounds centered.

Interesting point! I hadn't considered unequal response of ears but it makes sense.

If I do it by ear to get a mono signal to sound centered, would you have any idea what type of signal would make it easiest to hear the balance best? Sine wave at a certain frequency, white/pink noise, music with certain characteristics? I suppose I should just figure this out myself by trial and error, but is there any science-based (theory or evidence) answer to indicate what would best allow our ears to detect a left-right imbalance?

 

[fpitas]

Interesting point! I hadn't considered unequal response of ears but it makes sense.

If I do it by ear to get a mono signal to sound centered, would you have any idea what type of signal would make it easiest to hear the balance best? Sine wave at a certain frequency, white/pink noise, music with certain characteristics? I suppose I should just figure this out myself by trial and error, but is there any science-based (theory or evidence) answer to indicate what would best allow our ears to detect a left-right imbalance?

Pink noise is probably the most universal. The frequency response may vary between ears, as well as the sensitivity.

 

[Blumlein 88]

Most precise is play a 400 hz or so tone and set with the multimeter.

Next best would maybe be to play mono pink noise and adjust until it sounds centered.

With both tones or noise start with the level well down. Turn it up to something not terribly loud.

Worst would be with a microphone most likely.

 

[Sokel]

IIRC, it's a pretty nice Fluke meter, so I think it's likely to be true RMS, but I will check the model number when I get home.

For this method, would I just disconnect the speakers, play a test tone/noise, and measure AC voltage at the speaker terminals on the amp? I thought I read somewhere that it might be a bad idea to send a signal to the amp without speakers connected, but I can't remember why.

Should I use sine wave tones, pink noise, or white noise? If sine wave tones, should I be check at various frequencies or is one enough (e.g. @ 400Hz or 1kHz)?

You DON'T disconnect your speakers,you use a low level signal and be extra careful not to short the probes/inputs/outputs during the test.

 

[RayDunzl]

How would you suggest setting these gain knobs so that both channels are equal?

Audibly.

I used a pair of pots as a volume control between the CD Player and Amplifier for years.

The gear was centered between the speakers.

Mostly, I didn't change volume, just turn it on and go when I got home from the office or a road trip.

When I did change the volume, I adjusted the "balance" till it sounded right (if it wasn't right it didn't sound right) and went back to my seat.

Didn't worry about it past that.

Don't remember needing to fiddle with it after setting it (took what, 3 seconds?)

 

[fpitas]

I adjusted the "balance" till it sounded right

Another common control that disappeared.

 

[AnalogSteph]

For another method that hasn't been mentioned yet, place speakers directly face to face and play pink noise that's mono but with polarity inverted on one channel. (REW can generate that, for example, but it can also be done e.g. in Audacity.) Adjust balance for best cancellation, i.e. minimum remaining sound output.

As this method is subtractive, it allows for much better accuracy. Mind you, you may not even need it that accurate if there is any semblance of asymmetry in the room.

 

[RayDunzl]

Another common control that disappeared.

Well, not entirely, at least, not here in the present incarnation:

I have channel level at the miniDSP, the DEQ2496, the PC if using that, and the preamp, none of which I need to use, though the preamp is set to add 6dB to the unbalanced inputs to match the level of the balanced, in case they ever get used.

The pots are still in use in another room.

 

[Room_101]

For another method that hasn't been mentioned yet, place speakers directly face to face and play pink noise that's mono but with polarity inverted on one channel. (REW can generate that, for example, but it can also be done e.g. in Audacity.) Adjust balance for best cancellation, i.e. minimum remaining sound output.

As this method is subtractive, it allows for much better accuracy. Mind you, you may not even need it that accurate if there is any semblance of asymmetry in the room.

This seems like an ingenious method, but I'm not sure how well it would work for me, since I'm using dipolar planar speakers.

 

[DVDdoug]

The problem with pink noise and a meter is that noise is unstable. It's the nature of noise.

If you use a tone, the meter doesn't have to be "true RMS". The measurements just have to be stable and repeatable. And meter that's not true RMS sill reads sine waves correctly. Other waveshapes may not read correctly.

 

[tmtomh]

I'm a big fan of the centered mono image method, especially if you don't use EQ or DSP - because if you measure output at the speaker terminals and for some reason you feel the sound is not perfectly centered in-room, you're still going to have to fix that by tweaking the L or R gain knob on the amp. So why not "cut out the middleman" and just adjust them so a mono image sounds perfectly centered to you from your main listening position?

It's one of the few cases when I would say it doesn't really matter if issue is in the amp, the speakers, the room, or your ears - all that matters is that your image sounds centered.

 

[antcollinet]

Number 2 for me. Compensates for ear mismatch and room asymmetry.

Do you want to hear the sound centered between the speakers? Well adjust the balance until it is.

Simples.

 

[restorer-john]

IIRC, it's a pretty nice Fluke meter, so I think it's likely to be true RMS, but I will check the model number when I get home.

Use a low frequency, say 400Hz. And a sine. Depending on your model, the AC frequency response could be 1kHz max or up to 20kHz (with an accuracy hit) with the 87 series. The 1xx series will be reasonably accurate up to about 3kHz.

 

[ozzy9832001]

I'm assuming the knobs just turn and don't make a clicking noise or anything like that? If not, then I'd do it by ear with mono. Works well for me.

If they click, count the clicks from max.

If your room is asymmetrical (especially side wall distance) it's quite possible one will sound louder than the other volume matched.

In the case of asymmetry, a multimeter would be best that way your not corrected for the imbalance in the room and that can be done later on down the chain.

 

[Deleted member 48726]

If you don't use room correction, the best result would be to use your ears while playing mono music.