Excessive drumbeat when recording LP

[RHF]

Most specifically, Dave Grusin's Night-lines.

I have archived about 25 LPs to this point (headed toward 100), mostly from the 60's and 70's, so I am way too familiar with pops and clicks. When I record Night-lines I end up with excessive drum beat (I think) that look similar to really big clicks (too wide for repair, about 190-200 samples). I have compared the spectrums of the these beats and adjacent areas and most of the difference is much greater area at and above 5,000 Hz. I tried the low pass filter without much success. I can select the area and use a -30dB amplification and remove them or just cut them out. If I use a -15dB amplification I can reduce them to a "typical" drum beat that sounds similar to the recording when played on the turntable. However, there are literally 100's of them on each side. Does anyone have any suggestions to achieve similar results without having to select each beat and reduce its amplification?

Thanks,

RHF

Oh - Audacity 2.1.2 and OS X Yosemite 10.10.5

[kozikowski]

That's way too much work. You are producing damaged recordings. How are you recording the disks...connection details, part numbers.

Koz

[RHF]

OK, let's try this again.

To answer kozikowski's question: Ion Archive LP, USB cable that came with it, MacBook Pro (mid 2015).

I've attached (will attempt, anyway) two examples. The first is a 1972 LP (Ides of March - World Woven) and is an example of a "click" that makes it through the suggested work flow (click removal - a bit more sensitive than factory presets - threshold = 159, max spike width = 30; noise reduction at presets) and the second is the LP in question (Dave Grusin - Night-Lines - 1984). The first has a click that made it through processing just about in the middle that I listen for and remove using the Repair effect. The second has two "excessive drum beats?" at 0.172 and 0.632. The sound features are present in the original recording but appear to be selectively amplified during the recording process.

My question - Does anyone know of a way (using what I have on hand) to reduce this apparent amplification during recording or post-recording processing?

Both features are very visible in spectrogram view.

[Zebedee's Moustache]

It's going to take a genius to decipher what you're on about from two 1-second clips. It's difficult to know what you're doing wrong without more info, but above and beyond everything else the Ion Archive is a cheap piece of junk and you're never going to get decent sound from it no matter what you do.

[kozikowski]

I don't see any attachments or sound clips.

Prepare a ten second stereo clip featuring some normal music and some with the drum damage. We're going to compare them. As above, we can't do anything with one short, isolated, damaged drumbeat.

Prepare the WAV clip. Open a forum text window and scroll down to Upload attachment > Browse. A file dialog should appear. I do everything from my desktop, so my dialog always goes to the same place. Select your clip > Add the file.

Also:
http://forum.audacityteam.org/viewtopic ... 49&t=72887

Ion Archive

Did you used to have a really good turntable and is it still up in the attic? It may pay handsomely to repair that rather than struggling with the Ion. There are some good, inexpensive solutions to adapt a "raw," analog turntable to a computer. It's common knowledge you bought the Ion so you could put it in the trash. So that's what they built. Some Ions stop working sooner than they're supposed to, but they all have pretty poor pressure, tracking, skating and noise.

Koz

[cyrano]

I don't see any attachments or sound clips.

Me neither.

But they were there before. I downloaded them...

From the clips:

- No digital or analog clipping
- No apparent overload either, but I listened on laptop speakers. Sorry, didn't have anything else available.

It's not clear what the problem is. I first figured from reading the post that it would probably be poor tracking on the Ion's cartridge, but it doesn't look that way.

But the clips were a bit too short to tell, really.