How to map and EQ your dynamic cutterhead

[spinnertownblanks]

We have a blog on our website where we post 'hopefully' useful information for people learning to use a lathe and get into cutting. Our latest blog post has had a lot of clicks and thought it might be good to post in here too. This is advice from us specialising in the Vinyl Recorder, but the idea remains the same no matter what dynamic head you are using. Blog post below!

--------------------------------------------------------------------------------------------------------------------

As always, we have been actively providing support to cutters, with a recurring issue being the quality of their cuts, particularly when using the stock Vinyl Recorder setup. This is due to the Main Unit being inefficient using just a 3 band EQ to connect the heads response, when in reality it needs around 10 bands.

We consistently advise replacing all stock electronics, mapping the frequency response of your cutterhead, and using corrective EQ to flatten its natural curve while accounting for resonances. This approach ensures professional, flat cuts and maximises the system's potential. The VR is an amazing machine is used correctly, capable of great cuts.

For disc cutting, a professional setup typically adheres to the industry-standard RIAA specifications, which involves using an inverse RIAA curve before the cutterhead and the RIAA curve in the phono stage during playback. However, because Vinyl Recorder setups utilise tweeters instead of transducers in the cutterhead, the standard RIAA curve is not sufficient.

Another important factor to consider is the type of diamond used. A longer diamond, typically included with the machine, will produce more high-end frequencies, while a shorter diamond from other manufacturers will produce less. Both types are effective, but your final EQ curve will be tailored to the specific diamond type you are using.

Here’s a guide to doing it yourself. We will write this in a way that works with any Vinyl Recorder setup, no matter how your audio is routed. If you're unsure about the process, we also offer this as a service. Take your time, proceed methodically, and familiarise yourself with the intricacies of your machine.

Note: These instructions apply to all dynamic cutterheads, we just specialise in support for the VR so using it as our example.



Step 1: Make Sure Playback is Calibrated.



This step is critically important: you must ensure that playback is calibrated to the 0dB reference level using a professional test record. Do not use the stock test record that comes with the machine, as it is inaccurate. We recommend using the Ortofon Test Record. If calibration proves impossible due to the stock level meter having glued knobs, making it difficult to balance the left and right channels correctly, you may need to use an RCA attenuator or replace the level meter.

Step 2: Cut Audio Without EQ.



First, ensure you have appropriate audio for this task, which involves both program material and pink noise. The program material should be audio that is specifically mastered for vinyl and has a well-balanced frequency spectrum. This is crucial, as the quality and characteristics of the audio will directly impact the accuracy of the final EQ curve. The pink noise is simply used as a final reference to double check it is sitting well overall.

The next part will be cutting a disc, this has to be carried out with caution, if you accidentally send too much audio you could blow your fuses and damage your cutterhead. 
First you need to route your audio directly to the cutter amp, bypassing the Main Unit. To do this, unplug the audio inputs from the Main Unit and connect them to the cutter amp. The Main Unit's output, which previously connected to the cutter amp, is no longer needed and should be disconnected.

You are now ready to cut the audio. Start with your volume right down sending no signal at all, start to cut at 45rpm and monitor with your tonearm, then slowly raise the volume until cutting at +2dBu. This volume level is crucial, which is why accurate playback calibration is essential. Cutting at around +2dBu without corrective EQ will get the cutter head to exhibit its distortion profile and reveal its natural and mechanical resonances. We want to stress, this needs to be carried out carefully, exactly as we described and calibration MUST be right, if you calibrate with a stock VR test record you may easily blow your cutterhead.

Step 3: Record Audio

Next, record the audio back into your DAW as a 32-bit or 24-bit WAV/AIFF file. Keep in mind that the playback stylus can slightly influence the overall sound. If you don’t have access to a high-end stylus, we recommend using a widely-used model like the Ortofon 2M or Concorde series, as these are common and provide reliable performance. If you have access to the frequency response specifications of your playback stylus, you can use this information to adjust your final EQ, but it is not essential. The audio will sound muddy with a distinct lack of high end, this is normal.

Step 4: Make Your Custom EQ


For this process, we recommend using an EQ with a robust Match EQ function, unlimited bands, and various band shapes. We recommend either of these two plugins for this FabFilter Pro-Q or Toneboosters Equalizer Pro which provide all the necessary features.

Start by aligning your recording with the original audio in your DAW so you can A/B compare them side by side. Ensure both files play interchangeably and have matched volume levels. Since your recording might have little high-end information, adjust the low end by ear to balance it with the original, so both tracks' low ends are at similar volumes.

Next, use the EQ Match function with the original file as the reference. Create a 1-minute loop to ensure both files play identical audio into the analyser, then match the EQ.

Before cutting a new disc, apply the EQ to the original audio. To prevent digital clipping due to the EQ's high-end boost, normalise your audio to -23 LUFS. Apply this normalisation to all audio files to keep them at a consistent volume level that avoids overloading the plugin but is still capable of cutting up to +12dBu on playback.

For example, if you're working with a vinyl-mastered album, normalise the loudest track to -23 LUFS, then adjust the gain of all other tracks to the same amount. This will maintain the album’s intended overall volume and sound integrity.

Step 5: Check and Adjust Your EQ

Cut the audio and pink noise through the new EQ. Record these, then perform another EQ match in a new instance of your EQ plugin. Adjust as needed until the A/B comparison between your cut and the original file is as close as possible. If your original audio is mastered well and suitable for the task, along with the cutting volume being right to expose the heads resonances, it should be pretty much perfect.

Step 6: Done!

You should now have well-balanced cuts that significantly enhance your system's performance. If you are apprehensive to do this yourself, we offer this service through our store.

For an example of a cut on one of our VR heads compared to its original file, visit: https://www.dropbox.com/scl/fo/jdt2wdrk3ggblrbj4ckbx/h?rlkey=nlejue79f6y7w0ejpojb1l8yx&dl=0

[David34]

Thanks for this!

When you say 'dynamic head' will this work with the default one from Souri?

[markrob]

Hi,

This could work for any head. Its just a way to create an EQ calibration curve. How severe that is depends on the head.

Mark

[spinnertownblanks]

Thanks for this!

When you say 'dynamic head' will this work with the default one from Souri?

The VR head is what we are referring to as a dynamic head. As Mark said this will work with any head, but generally feedback systems will have a preamp with better EQ. If you have any questions about it always feel free to reach out, happy to help.

[Fattcamp]

Hey there,

First off, thanks so much for this detailed instruction. I've been using this guide (with Fabfilter Pro-Q 3) to EQ my 3D-Printed lockdown lathe cutterhead. I had a couple of questions as this is my first foray into record lathes/cutting (embossing in my case). I'll just sort of describe my experience going through these steps with some questions along the way. All embossing done at 45 RPM's with Mike's Sapphire embossing cone.


1. Calibration: I had already (physically) calibrated my turntable prior but bought the Ortofon test record either way as per the instruction. When you say calibrate to the 0db reference level, are you referring to ensuring the tracks on the test record play back as intended or specifically calibrate the meter? I'm in logic and doing any metering in the box. I'm currently using a Stanton STR8.150 with the line-out into my interface as opposed to hardware metering. It probably sounds like a dumb question but since everything is based off of this initial calibration, I just want to understand properly.

2. Cutting without EQ: I mono-ized a stereo track to use as a reference and cut it as loud as realistically possible without distortion and still being able to play back on the TT (This track had to be reduced in volume quite a bit for this step). This seemed to work fine. I'm not sure how to ensure cutting around +2dbu though. Faders in Logic are dBFS from what I could find. I'm playing audio to be cut using David's Vinyl Burn software, which is sent out from my interface to an ART SLA-2 power amp (200 watts/side @ 8-ohms) to my cutterhead which uses 8-ohm Tectonic TEAX-19C01-8 coin exciters as per the build.

3. Recording Audio: Recording of audio seemed fine and and lacked high-end as expected

4. Custom EQ: I used Pro-Q 3 match-eq with the un-effected reference track level matched & side-chained to the recorded audio in step 3. It was definitely in the ball-park but you could hear quite a bit of high frequency noise. I then applied this custom EQ to the un-affected original track and attenuated the volume to avoid clipping. Again, with no physical meter and working ITB, I'm not sure how to ensure it's cutting up to +12dbu.

5. I then listened and recorded the custom eq'd track (mono-ized) into the DAW. I no longer had the high frequency harshness from when creating the curve which was nice. It's still not quite there yet but the audio was definitely in the ball-park of the original track albeit with less highs (I think this is somewhat expected with embossing on polycarbonate). I decided to try a second instance of Pro-Q 3's match-eq to get even closer to the original. This second custom EQ further bumped high end quite a bit. This is where I got a bit confused on how to apply, so I added 2 instances of Pro-Q3 on the original un-effected track, one with the first custom EQ, the second instance with the second custom EQ. This led me to having to attenuate much more before cutting due to the huge amount of top end. I've avoided applying HF limiters or De-Essing yet as I figured we wanted to keep the cutterhead calibration process as raw and un-effected as possible. After cutting again and re-recording the dual-custom eq'd track back into my DAW, it was surprisingly muddier and lacking top end. I don't understand this... but I know I had 2 x fairly extremely boosted custom EQ's stacked on top of each other... and had to cut quieter. The result was quite distorted as well. Nothing clipped digitally while applying the dual EQ, nor was my amp clipping. I guess it was just too much high end? I also tried the whole dual EQ process again adding gain-reduction between the Pro-Q 3 instances and same result. Am I hitting an inevitable wall for what can be boosted and expected back via embossing? When creating the second custom-EQ ITB, it was VERY close to the original without the distortion... but stacking the dual-eq on the original & cutting resulted in the strangely darker top end and added distortion.

-----------------------------

Sorry for the long read... I just felt like sharing my experience. Currently, I prefer the sound of the single custom-EQ but would love to be able to squeeze a bit more top end out of it if possible. I'm wondering, should I be turning down the track volume and cranking the amp more (would add noise I imagine), maybe I'm missing something simple here. I know we need the extra power for these tiny drivers but not sure if the amp volume itself is part of my issue since the signal being fed to it is plenty strong. I'm running my amp at 12:00 for both channels (-18dB) and and its led's are showing signal without clipping while cutting.

I guess there's also the possibility of cutting half-speed to try and help with the top end, but then I won't be able to fit as much content per-side and I'm trying to fit a 42-minute album on a 12" disk.. but I guess I can always half-speed it over 2 discs. Would I need to re-calibrate the head for half-speed cutting?

If anyone is still with me, I'm all ears for things to try. I can manually of course lower the highs on the second custom-EQ but I liked the result ITB adding the second custom EQ before actually cutting with it.

I've attached audio showing the various results:

Track 1 - Original audio (Mono version of original stereo track)

1. Original Audio.mp3

Track 2 - Full range cut (Lacks highs as expected)

2. Full Range Cut (No EQ).mp3

Track 3 - Match EQ Cut (Not bad)

3 . Match EQ Cut.mp3

Track 3.1 - Dual Match EQ (This File is a bounce from my Daw (Logic) of the Match EQ cut (Track 3) but with second match eq added. Much closer to original but not cut yet)

3.1 Dual Match EQ (Same Match EQ cut bounced from DAW with second match eq).mp3

Track 4 - Dual Match EQ Cut (both match eq instances applied)

4. Dual Match EQ Cut.mp3

Thanks,
Dave

[Big Al]

Hi @dave

I think if you want to sound close to real records then maybe move from embossing in mono to actual cutting in stereo. I think this really is the goal to the sound of what great records are! Definitely not easy but achievable i.e. Souri’s stereo lathe / head - what most ‘pro’ ‘Lathe cutters’ use, which when used by someone who know’s what they are doing can (depending on the genre / style of music & production quality of what they are cutting) can sound pretty damn close to a ‘professional’ pressed record (most likely cut on a Neumann SX etc., ) trying to emboss in mono is not going to sound in the same ballpark and will be generally perceived as more ‘lo fi’ . Just having a nice stereo image of sound is perceived as being more ‘Hi fi’ and ‘brighter’ than mono in most styles of modern music! Fair enough if you produce 70’s dub for example which is mostly mono but for modern music stereo all the way baby! Hope this helps! Took me 8 years to find the unicorn of Japanese small ‘portable’ lathes here in Tokyo that is actually ‘STEREO’ (as most are MONO) this Hara machine actually has two of the usual heads you find in the mono machines, setup for stereo and the sound of the disks cut (on Myshank blanks using his ACE diamond Styli) is so much better quality then mono embossing could ever achieve in my experience!
Hope this helps in your quest for nice sounding records! To sum up - try to aim for ‘Cutting’ with a Diamond in stereo as this will probably more likely be the sound !

[Fattcamp]

Hey Big Al,

Thanks so much for taking the time to read and respond. I totally hear you on cutting stereo with diamond styli. That may be the path I ultimately take, but after researching this forum, I chose the more cost-effective (still not cheap!) option of building my own lathe with 3d-printed parts and using Mike's sapphire cone on PC.

I wasn't into the idea initially of the more expensive diamond styli that apparently don't last as long... even though the results would be better. This is more a hobby of mine to see if I could even make it this far... but now I've accomplished my goal, now I'm and wanting to squeeze as much fidelity out of it as possible keeping in mind my system/format limitations. I'd like to maybe do some small runs for artists locally... maybe some Etsy one-offs or something like that, but it's been all about the fun of getting to this point. It's been a blast!

The other thing, is I'd need to get a vacuum system set up and add heat to the diamond stylus, and seeing as my cutterhead is currently 3d printed in PLA+, I'd probably need to get it laser-sintered in alumide like David's original build (or perhaps at least the torque tube)... so it would be a decent chunk of $$$ to get there. I'd love to pick up one of Souri's T560's or something capable of stereo at some point, but it's just not realistic at this time. I think you can retrofit stereo heads on Presto's which would be fun too. I guess Mike would be the guy for that sort of thing but I haven't gone down that path yet.

That Hara lathe you've got sounds really great! I haven't heard of them but I'm new to all of this. I'm glad you were able to find a stereo cutting rig that gives the results you were after! I'm going to keep calibrating my mono rig using various reference tracks and see what I can get out of it but I'm happy to be where I am... for now!

Thanks,
Dave

[Big Al]

Hey Dave hope it goes well! Yes good to start off with a mono lo fi rig, that’s what I did and learnt the foundation of cutting, embossing was a challenge as it seems to be more lo fi than even cutting in mono and way more challenging to get to play back without skipping than cutting but its probably the cheapest option to get going but because of the extra challenges it helps to learn about preparing the source material to get the best results which is a good skill to learn when cutting!
Good luck with getting a Souri lathe if you ever go dowm that path, not an easy feat!

[spinnertownblanks]

Hey!

Detailed analysis is always welcome, good work. To summarise:

Our post was specific to stereo diamond cutting like Big Al mentioned, maybe we should have made that clearer, we have no experience with embossing but the theory for the cutterhead is the same, but fidelity will likely be lower. I havent had a chance to check your recordings yet but will do later.

On a side note, if you are interested in diamond cutting, very soon a new lathe will be on the market, we have heard test cuts with a properly EQ'd head using the method above and seen how the lathe works, and it comes with a very good price tag. We can't give more info yet as it is not our project, but they are in use and in our option sounds incredible, it will hold up next to a VR T560 in terms of sound and a much better price, even when new. So this will open more opportunities for hobbiest's like yourself.

[Fattcamp]

Hey!

Detailed analysis is always welcome, good work. To summarise:

Our post was specific to stereo diamond cutting like Big Al mentioned, maybe we should have made that clearer, we have no experience with embossing but the theory for the cutterhead is the same, but fidelity will likely be lower. I havent had a chance to check your recordings yet but will do later.

On a side note, if you are interested in diamond cutting, very soon a new lathe will be on the market, we have heard test cuts with a properly EQ'd head using the method above and seen how the lathe works, and it comes with a very good price tag. We can't give more info yet as it is not our project, but they are in use and in our option sounds incredible, it will hold up next to a VR T560 in terms of sound and a much better price, even when new. So this will open more opportunities for hobbiest's like yourself.

Thanks for the clarification. As it turns out, your process works quite well for my mono embossing setup. I think I'm just hearing the limitations of what can be done with my rig. I also found out that after doing a bunch of test cuts, I started to dull the sapphire cone and needed to rotate it 90 degrees.... so I guess if I go too nuts trying to calibrate it, the very wear will affect said calibration. More tests to come.

Sounds VERY interesting on this new lathe that's in the works. I'll keep an eye out for it!

Dave

[markrob]

Hi,

I've found that when embossing there is a high freq limit based on serval factors. They include the embossing stylus geometry, platter speed, and cutting level, drive amp peak instantaneous power, and cutter head design limits. You might think that on playback you can bring up the highs with a EQ and then apply that extra correction on your next cut and end up with the response you desire, but the self erasure of the embosser wipes out the extra EQ and just ends up being a waste of power to the head. You might be able to get closer to the response if you cut at a lower level, but since embossing already requires trade offs on cutting level its probably not worth it. If you think the limit is due to lack of amp power and/or head design limit, half speed mastering will work well. There are probably some better embossing stylus geometries (e.g. the EPOS style embosser looks very good), but if you make the embosser contact area too small you will end up transition from embossing to cutting with a very noisy result. Hope this makes some sense.

Mark

[Fattcamp]

Hey Mark,

Great info here. Appreciate it. Yeah it's interesting how easy it is to match EQ digitally, but then immediately hear that it cannot necessarily be replicated to a physical disc. The first instance of match-eq got it in the ballpark and I've found that I can take the second instance, but reduce the large 20db(ish) bump in the highs and reduce it to a 2db bump or so and I can squeeze out a bit more top end without distortion... but it's still not going to get all the way there (nor can that be expected). It's just interesting to use the tools available to see how far you can take it.

I mentioned the amp, because it's 200-watts a side (8ohms) and some posts here suggest a bit more... but some have used less with decent results. I wasn't sure if the fact that my amp is clearly showing it's getting decent signal... but is only at half of it's max volume somehow made a difference (as opposed to lowering the volume on the track and cranking the amp... but then we'd just be introducing more noise). I liked the easy visual of keeping the amp volume knobs at 12:00 for consistency but maybe I'm not approaching that step properly... or maybe lacking a bit of power?

As far as half speed goes, I think the closest I can go with my setup is slow down the digital track to equivalent of 33 1/3 RPM's in Logic using varispeed (speed + pitch down -25.93%) and then cut at 33 1/3 in order to play back at 45 RPM... but then have to wrap my head around which version/speed of the digital files I'm using the match-eq with. That's a bit of a brain scrambler for calibration.

Dave

[markrob]

Hi Dave,

33 for 45 is a great way to go. It does get tricky to deal with the EQ. One way to handle that is to calibrate using no RIAA and a flat preamp for playback. I have a mod for a off the shelf BBE FJB-200 unit if you are interested. This will let you find an EQ to flatten the open loop head response. You can use this EQ at lower speed since the head response will not change, just lower overall bandwidth. Then, apply your IRIAA as a secondary EQ with the turnover frequencies (50, 500, and 2122hz) scaled by the speed (multiply each frequency by the factor 33/45). You could do that by applying the IRIAA as a plugin to your source digital file at full speed and then pitching the result down to 33. Finally, add a steep (maybe -12-18 db/oct) low pass at the tail end to limit the bandwidth fed to the head based on the speed. So if you cut 33 for 45 and are looking to 12Khz bandwidth on playback, low pass at 8.8Khz (12 *(33/45). That could be done to the pitched down sample prior to cutting.

This all maybe outside the scope of the original post. I didn't mean to hijack it.

Mark

[Fattcamp]

Hi Dave,

33 for 45 is a great way to go. It does get tricky to deal with the EQ. One way to handle that is to calibrate using no RIAA and a flat preamp for playback. I have a mod for a off the shelf BBE FJB-200 unit if you are interested. This will let you find an EQ to flatten the open loop head response. You can use this EQ at lower speed since the head response will not change, just lower overall bandwidth. Then, apply your IRIAA as a secondary EQ with the turnover frequencies (50, 500, and 2122hz) scaled by the speed (multiply each frequency by the factor 33/45). You could do that by applying the IRIAA as a plugin to your source digital file at full speed and then pitching the result down to 33. Finally, add a steep (maybe -12-18 db/oct) low pass at the tail end to limit the bandwidth fed to the head based on the speed. So if you cut 33 for 45 and are looking to 12Khz bandwidth on playback, low pass at 8.8Khz (12 *(33/45). That could be done to the pitched down sample prior to cutting.

This all maybe outside the scope of the original post. I didn't mean to hijack it.

Mark

Thanks markrob for the breakdown on this. I think I'll DM you and see if it's worth another post on this. Definitely curious on lower speed embossing and a to get a bit more info on your end... plus combining with these eq steps. Sorry spinnertownblanks! Your EQ instructions are great and opening up other combinations/possibilities.

Dave

[spinnertownblanks]

the whole point is to share information and help each other, so this is great

[Fattcamp]

the whole point is to share information and help each other, so this is great

Sounds good!

Hi Dave,

33 for 45 is a great way to go. It does get tricky to deal with the EQ. One way to handle that is to calibrate using no RIAA and a flat preamp for playback. I have a mod for a off the shelf BBE FJB-200 unit if you are interested. This will let you find an EQ to flatten the open loop head response. You can use this EQ at lower speed since the head response will not change, just lower overall bandwidth. Then, apply your IRIAA as a secondary EQ with the turnover frequencies (50, 500, and 2122hz) scaled by the speed (multiply each frequency by the factor 33/45). You could do that by applying the IRIAA as a plugin to your source digital file at full speed and then pitching the result down to 33. Finally, add a steep (maybe -12-18 db/oct) low pass at the tail end to limit the bandwidth fed to the head based on the speed. So if you cut 33 for 45 and are looking to 12Khz bandwidth on playback, low pass at 8.8Khz (12 *(33/45). That could be done to the pitched down sample prior to cutting.

This all maybe outside the scope of the original post. I didn't mean to hijack it.

Mark

In my particular case, my TT is going direct line-out to my interface for playback using the stock built-in RIAA. I'm hoping to use this spinnertownblanks map/eq method and was trying to understand which version/speed of the cut and recorded files I should be using. If I'm understanding correctly, your method of the modded phono pre takes the RIAA out of the equation for playback.

I was hoping to be able to play with Logic's varispeed, slow down the audio appropriately, and just determine whether I should be trying my match-eq to the full range cut (Step 2) using the slowed audio or recorded back into my DAW at regular pitch/speed (running at 45 RPM's) if that makes sense.

My first attempt gave me quite different results than I was expecting. I slowed the audio & pitch in my DAW, cut at 33 1/3 RPM, recorded it back in at proper speed/pitch (45 RPM) and performed the EQ match... therefore matching the proper time/pitch with the full range cut and original audio... but I'm wondering if I should have match-eq'd to the pitched down versions in both cases. I got a much different tone as everything felt shifted a bit.

Perhaps I need to separate the bounce of the slowed audio from the custom EQ. I'm just not quite seeing the light on how best to approach this. I'll re-read the suggestion of using the phono pre to try and understand it better but was trying to avoid yet another purchase for this lathe... haha! I needs me some center labels and blanks.

Dave

[markrob]

Hi Dave,

You could pass your program material through a standard IRIAA only and record at normal speed. Use the match EQ function to flatten the response of the head as best you can. Don't try to flatten much past 10Khz or so as this range of the head response will be unused at the slower mastering speed. This is EQ is a first approximation of your head's open loop response since you pre applied the IRIAA to your program material and you phono preamp removed it on playback. Now create a pitch scaled (see my last post) IRIAA EQ and apply this to your source after you pitch down the program material to 33. Next, apply the EQ you matched previously. I would add a steep low pass to remove info that will be outside of your desired bandwidth (see my last post). That should work without having to go the flat preamp route. Just be careful on levels since you will be boosting the highs via the IRIAA. Depending on your head response, that might be a good thing as the IRIAA boost will pull the high frequency response out of the mud.

Mark

[Fattcamp]

Hi Dave,

You could pass your program material through a standard IRIAA only and record at normal speed. Use the match EQ function to flatten the response of the head as best you can. Don't try to flatten much past 10Khz or so as this range of the head response will be unused at the slower mastering speed. This is EQ is a first approximation of your head's open loop response since you pre applied the IRIAA to your program material and you phono preamp removed it on playback. Now create a pitch scaled (see my last post) IRIAA EQ and apply this to your source after you pitch down the program material to 33. Next, apply the EQ you matched previously. I would add a steep low pass to remove info that will be outside of your desired bandwidth (see my last post). That should work without having to go the flat preamp route. Just be careful on levels since you will be boosting the highs via the IRIAA. Depending on your head response, that might be a good thing as the IRIAA boost will pull the high frequency response out of the mud.

Mark

Hey There. Appreciate the info here. I don't have an EQ plugin that has RIAA or IRIAA so I've just been passing my original audio to the cutterhead (Standard RIAA from Turntable / line out) and trying to flatten the best I can from there using the original instructions. Apparently Audacity has an IRIAA preset and I didn't realize they have a mac version since I only use Logic. Maybe I'll try that unless you know of any mac eq plugins that contain IRIAA presets.

In the meantime, I tried another 33-1/3 for 45 test and the results are below. Last time I tried this, I applied the match-eq and then pitched down the material to cut at 33-1/3... and once recording back into my DAW at 45 RPM (regular speed/pitch), the sound was quite off (see above post/results). This time I pitched down the audio first, cut at 33-1/3 full range without any EQ, recorded into DAW at 33-1/3 to maintain lower speed/pitch and created match-eq in that pitched down state. Lastly, I ran the original (pitched down) audio through the new match-eq (created while pitched down) and cut at 33-1/3. I recorded back into the DAW at 45 RPM and the results seem better in this case. I think this was part of your steps above about pitching down first and then creating a match-eq but maybe I didn't quite understand your instructions.

Let me know what you think of File #2 (Standard 45 RPM cut with match-eq done at regular speed) and File #3 (33-1/3 for 45 RPM with match-eq created to pitched down audio). It seems to have helped in the mid-range and has a bit of a "less hollow" sound and helped in the top end a bit. When listening back to these new tests in my DAW, I cut off some highs from the original audio to see which of the 2 results were closer and appears to be the 33-1/3 for 45 RPM file #3 is.

Thanks,
Dave

*** Sorry for the slight volume differences. The meter showed relative levels the same ***
*** Both results have no HF Limiting while I sort out the best way to go ***

Original Audio

1. Original Audio (Test 2).mp3

Match-EQ (45 RPM)

2. Match EQ Cut (45 RPM - Test 2).mp3

Match-EQ (33-1/3 for 45 RPM)

3. Match EQ Cut (33 to 45 RPM - Test 2).mp3

[markrob]

Hi,

Looks like you did a pretty good job. I compared the two cuts to the original using Har-Bal and also listening. I agree the 33 for 45 sounds better and if you look at the two screen grabs, matches better. The high frequency is probably due to embossing vs cutting. You might be able to get closer if you cut at a much lower level.

Here is the 45 cut comparison:

45.png

and the 33 for 45 version

33.png

[Fattcamp]

Hi,

Looks like you did a pretty good job. I compared the two cuts to the original using Har-Bal and also listening. I agree the 33 for 45 sounds better and if you look at the two screen grabs, matches better. The high frequency is probably due to embossing vs cutting. You might be able to get closer if you cut at a much lower level.

Here is the 45 cut comparison:
45.png

and the 33 for 45 version
33.png

Awesome! Thanks for the visuals & feedback on these cuts. As a total noob with my first DIY 3D-printed mono embossing rig, I'm happy with this latest 33 for 45 result. I wish I could cut slower, but I'm currently using a stock TT that I can't run lower than 33-1/3. The Stanton STR8.150 seemed to be a good high-torque option without having to build one but building or acquiring something that can cut slower could improve things even more.. unless you know some magical way to cut slower with the TT speed restriction!

I wonder if I can modify the programming of the arduino and how it communicates with the linear rail motor... but that's only adjusting what's sent to the cutterhead as opposed to the platter speed itself. I'm just thinking out loud now.

My next test is the same 33 for 45 but with a second EQ (or modified match-eq) adding a HPF cutting off anything under 25hz which will have to be multiplied by a factor of (33/45) I guess to make it closer to 18hz. I'm also going to finally add a HF limiter plugin instance. So far I'm playing with Airwindows "Acceleration2" which seems to be easy to use, subtle but does what I need. There are lots of other ways to do this. Not sure what you guys like to use but this next test should be about as good as I can treat audio for my rig without further hardware upgrades.

Do you guys like your HF Limiting after any EQ-matching or before? I guess it would be more subtle or less effective before extreme boosts in high frequencies compared to after but just curious.

Thanks again,
Dave