Advice sought; EQ following speed conversion of tape

Hi. I’m new here. I wonder if someone can help with this …

I’m transferring a range of 3.75 ips tapes for restoration. Rather than using a geriatric Akai which will only reluctantly do 3.75 ips, I’ve been using a B77 Revox at 7.5 ips and using Audacity to do a sample rate conversion to the correct speed. This has been very successful as far as it goes. However, my limited understanding of the subject tells me that my NAB EQing will be askew.

My question is; is there a methodology using Audacity (as there is RIAA with 78 vinyl) in that you can re-set the curve by first inverting and then reapplying a correct curve? Or have I got this all wrong and it’s just done by ear?

Any advice appreciated

This has been very successful as far as it goes.

Then I should keep doing it that way. There is no direct translation between tape speeds because it’s not just pitch and equalization. The head resonance adjustments and velocity correction will also be off. That creaking sound is you opening a can of worms. If you have a really good sound system or good headphones and you’re pleased with the work, then go for it, as is.

Playing 78s on an LP stereo system doesn’t work very well. It’s desperation method, not transfer technique. An LP needle will hit the bottom of the wider 78 groove and pick up noise and distortion that were never part of the original performance. If you do manage to get the needle right, you need a music system that goes down in pitch beyond audibility to get the 78 original frequency response.

And the RIAA curve.


I presume that these are old tapes, so the sound will have deteriorated to some extent, so the “correct” NAB Eq will probably not be the optimal Eq any more. Better to Eq by ear (if it needs Eq at all).

If your sound-card has the ability to record at a higher sample rate, say 96kHz rather than the typical 44.1kHz, you should try using the higher rate. If you record at a sample rate of 44.1kHz the conversion method you’ve described will lose everything above 11kHz which is on the tape, because you are halving the playback-speed.
[ 11kHz bandwidth may be adequate for your purposes, but most people can hear up to about 14kHz , youngsters can hear up to 18kHz or higher ].

Even using the higher sample-rate there may still be a 11kHz limit because the analogue electronics in the tape player and sound-card are designed for audible sound, ( i.e. below 20kHz) , doubling the tape speed [to 7.5 ips] will cause some of the high-frequency audio to be shifted into inaudible ultrasound (above 20kHz).

Thanks guys.
My process thus far has been to drop from the B77 though a RME UCX to Audacity at 88.2kHz. The top end of the Revox is 20 kHz +2 / -3 dB at 7.5 ips (when new). I then set to 44.1.

I hadn’t considered the analogue limitations in my thinking. In this instance, the bulk of the tapes are solo piano from the 80’s and some I’ve had to bake.I had thought there may be some conventional wisdom or time-honored methodology to adopt, but I don’t know how the tapes where recorded in the first place, I guess my question, on further reflection, was only half-baked itself. I have RX3 to strip and a bunch of András Schiff CD’s to EQ to so I should be OK.

Thanks again for your help.

P.S.Koz: I learned 78 transfer the painful way. These days I use a modified AT LP120 with a 78 cart and a range of after-market wide styli.

If you then look at the a spectrogram-view of the sound in audacity , you will readily see if any piano above 11kHz has survived the transition to digital via your double-speed playback method.

Thanks Trebor.
The answer is that not a skerrick above 11k has survived the voyage. But for this exercise, it’s not tragic as I guess it’s mostly hiss above 8k anyway. Still, it sounds better than a transfer from my old Akai M-8 with all it’s wowing and fluttering. Maybe I should try and find a low speed B77. Anyone here in New Zealand with a spare?
Thanks all for your great advice.