stereo extraction from Edison cylinders

[Dan McDonald]

With the old mono acoustic 78s, I have had good luck extracting a stereo wav file, then using various addition or subtraction strategies to eliminate the noise that doesn't appear in both channels.

I had the idea that I should modify the device I use for extracting the sound for cylinders, but just wanted to see if anyone else has done this for cylinders or not.

Before I go to all that trouble, I was just wanted to make sure it would work as well as it does for 78s. Has anyone tried this?

Dan McDonald

[Dan McDonald]

Re: stereo extraction from Edison cylinders

Well, I went ahead and did - after about 3 hours of work, I got it together and results so far are good. Subtracting l-r gives a fairly clean signal - cleaner than I got with a mono signal, anyway, so I'm happy.

Dan

[Craig Maier]

Re: stereo extraction from Edison cylinders

I am not so sure that I understand why you needed to modify the machine. What am I missing - - - the Diamond Cut software has L-R as a file conversion option. It also has L+R and L only and R only. Couple that with the paste mix routine and I think that you have it. Could what you have done been achieved using these routines?

[Dan McDonald]

Re: stereo extraction from Edison cylinders

Hey Craig -

Well, the device I use to extract from the cylinders produces a mono wave file, and doing l+r tends to make the noise relatively weak compared to the signal, but if you extract it in stereo you get two different sets of noises, so when you combine with l+r or l-r you are able to get rid of sound that is not in one of the channels (which should be pure noise, right?).

This way seems to get rid of a lot of the clicks that are a result of a piece of dirt or a scratch that hit one side of the channel only.

At least I think that's what it's doing.

Dan

[Dan McDonald]

Re: stereo extraction from Edison cylinders

Right, but with cylinders you have to have a stereo phono cartridge - that is what I was asking about. I guess it wasn't clear from my post, but I had to figure out how to do Edison cylinders with a stereo phono cartridge. I was just checking to make sure my thinking was correct; I do pretty much what you do with Diamond Discs, and wanted to try it with cylinders.

Thanks Craig,
Dan

[Craig Maier]

Re: stereo extraction from Edison cylinders

Oh, yeah!! Now I see. Of course. It took three hour to rig up the Stereo phono cartridge to work on your cylinder player. Actually, that is pretty quick for a modification like that. Good work! Maybe you could tell others how you did it - - - sounds like an interesting weekend project.

[Craig Maier]

Re: stereo extraction from Edison cylinders

They way I deal with Edison recordings which are Hill and Dale (vertically modulated) is to record them using a stereo phono cartridge onto my hard drive. Then to extract the vertical component of the signal (the component with the least amount of noise, usually) and then apply the file conversion filter with it set to L-R. Futher manipulations could be done using the Edit paste functions if desired. But, no special hardware should be required to do this job. All the tools are in the software to make the process relatively easy.

[Dan McDonald]

Re: stereo extraction from Edison cylinders

Thanks, Craig. Well, the three hours was just to modify what I already had, which was the ACT (Advanced Cylinder Technology) device for recording from cylinders. If you haven't seen it, it's a small piece very similar to a reproducer but it has a Sure 500 phono cartridge inside instead of the original Edison needle.

All I had to do was re-wire the thing, but the space is so cramped that that is why it took the three hours. There really isn't much space - you have to solder the wires directly to the pins on the cartridge, so it's a little delicate. I imagine that is why they manufacture them in mono. They use a stereo cartridge but just wire it as mono.

Dan

[Dan McDonald]

Re: stereo extraction from Edison cylinders

Craig -
One more question -

Does the Forensics Adaptive Filter setting using the right channel as a reference do essentially the same thing?
I tried it with a couple of acoustic 78s and got pretty nice results.

Dan