Can mismatched impedance degrade overall volume?

[SFDave]

Let me start by saying my setup is unusual, so bear with me. Instead of using studio microphones to record an acoustic grand piano, I'm using the Helpinstill stereo pickups (created for use in live concerts where acoustic pianos are on stage with electronic instruments and feedback is a problem, i.e. Billy Joel and Elton John both use this system). These are magnetic pickups and function the same way guitar pickups work. Unlike mics, the output is initially unamplified. As you would expect, the sound is very dry and needs processing to add ambience - reverb etc, all of which is done in LogicStudio. It's only been recently that this system has begun to be used for recording purposes, and I may be one of only a few people trying to go direct to digital with it. Though it's not really any different than recording an electric guitar directly to computer.

The controller box has both balanced studio type outputs (low impedance?), and optional regular unbalanced large phono outputs. I am using the large size phone plugs with adaptors to reduce the two mono lines to a single mini stereo plug which goes into the mic input of my Mac Powerbook running variously, GarageBand or Logic Studio, and I'm sure the computer input impedances don't match the pickup's output.

My problem is that the signal is apparently recorded OK, volume levels look OK - when I view the waveform in an editor, it shows up as a nice fat wave barely contained within the clipping barriers. However, playback has a very low volume as if the piano were in another room with the door closed. Using a waveform editor to "amplify" the soundwave only makes for some nasty clipping, and the volume isn't any better.

My question is, is this a problem with impedance mismatching? I'm aware that mismatched impedance can cause degradation of the higher frequencies and add distortion or noise, but the overall volume is my issue. Could using a simple pre-amp solve the problem, or would that just make the same problem worse by adding noise? Is it worth a couple hundred dollars to add something like the Radial Engineering JPC Stereo PC DI Box into the mix? Or would the Yamaha AUDIOGRAM6 which includes a pre-amp and, I think, performs impedance balancing as well be a good idea?

Sorry, I'm not a sound engineer, so I'm hoping someone will understand my problem and can help find the right solution before I spend a bunch of money on the wrong one. Thanks!

Dave

 

[bruiser1964]

Firstly, the internal souncard in your comp is pants, they are just not spec'd for this purpose. You need an interface either firewire or usb (2 channel) for what you are doing, and these can be bought for around $200.
Going from balanced to unbalanced or using cable adapters is never the way forward.
Regarding what you were saying it sounds like you have a captured sound at reasonable volume but it won't output - or - more likely, most of the captured sound is hiss / distortion.
Good luck
PPPPS would mic's not be better? I've heard good results on piano with boundary mic's (cheap)

 

[SFDave]

... You need an interface either firewire or usb (2 channel) for what you are doing, and these can be bought for around $200.
Going from balanced to unbalanced or using cable adapters is never the way forward.
Regarding what you were saying it sounds like you have a captured sound at reasonable volume but it won't output - or - more likely, most of the captured sound is hiss / distortion.
Good luck
PPPPS would mic's not be better? I've heard good results on piano with boundary mic's (cheap)

It does output, only the volume is at about half of what it should be, even with the input gain control in the software all the way up when recording, and there is almost zero hiss/distortion generated by the unamplified pickups. I have a couple of good consumer grade Senheiser low noise studio mics (expensive, they were) that I've used in the past going through a small 4 channel mixing board that has pre-amps and pans. The sound quality was pretty accurate and free of radio noise, but (as with any acoustic mic) the problem of distortion arose because of certain harmonics in the acoustic piano that react with the mic and cause huge spikes in the wave - forcing me to record at too low a volume level to be acceptable. As you suggest, this may also be partly blamed on the consumer grade sound card in the computer. A good inline compressor might have helped some, but you can't compress acoustic piano very much without muting the overall quality of the sound.

Professionals all agree that recording acoustic piano is the most challenging in the business. I've also tried several other recording systems - one, a free standing (and very expensive) digital system made especially for acoustic piano (by Kawai, my piano manufacturer and loaned to me by my piano tuner) that used mics attached to the bottom of the piano. But it really sounded like crap - not even as good as the Senheisers and my computer's own hardware.

The real reason I'm using the pickups though, is that I am recording with my piano at home where it's almost impossible to get a clean recording free of ambient noise - i.e. the dogs barking, the bus going by on the street, fingernails on the keyboard or my foot squeaking or thumping on the pedal, not to mention the frustration of having a good 10 minute take of a Rachmaninoff piece ruined by one of the above.

Can you recommend a good interface? Is the Yamaha I mentioned in the first post a possible choice? I just found it on the internet and can't find any reviews about it. The Radial Engineering unit is only for balancing impedance and doesn't have pre-amps...

 

[sjjohnston]

Focusing just on one thing, which isn't even what you asked about:

If you see a big fat waveform, and you hear a low volume, something weird is going on that (probably) has nothing to do with impedance matching. The meter is telling you that you've recorded a signal of the appropriate amplitude (nearly full-scale, digitally) ... this suggests that failure to record a signal of appropriate amplitude isn't your problem.

Thoughts:
- You mention you're recording two signals with a stereo plug ... are they inverted (either exactly or approximately) versions of one another?
- When you play back other recordings that look right on the waveform display, do they play at the volume you expect? I'm guessing that's the case, but it's worth asking.
- There are a few ways you might get a recording whose waveform looks close to full-scale, but doesn't play back that way, but the ones that pop to mind seem unlikely. Examples:
* Huge DC offset (which would get you a signal that's close to clipping on the top or bottom, but not both).
* A signal that's primarily or entirely out of the normal frequency range, so that you either can't hear it or your playback system won't reproduce it. You could, for example, record a 20k sine wave at full scale, which would be inaudible (to normal humans, anyway) on playback. Seems unlikely, unless your pickup system is broken.

 

[sjjohnston]

One possibly useful test: record a single, sustained note somewhere toward the low end of keyboard, hitting the key with moderate attack and letting it sustain. Zoom in enough so you can actually look at the waveform in mid-note. It should look like clean, smooth wave: likely something like a sine wave with a little wiggle in the middle.

 

[bouldersoundguy]

- You mention you're recording two signals with a stereo plug ... are they inverted (either exactly or approximately) versions of one another?

This was my first thought. If the two pickups are getting summed to mono with mismatched polarity the two signals will largely cancel out, leaving only some ambient sounding highs. If they are summed before recording then the recording will be no good. If they are still separate as two mono files or a stereo file then the recording may be recoverable.

Is the recording 2-channel or mono? If it's 2-channel, what happens if you play back just one channel?