2022_06 The positioning of the cutterhead with cross laser
To ensure that the cutterhead is also positioned exactly right, the installation height must be correct (needle must have the correct angle when in contact with the record).
If the installation height is too high, then the cutterhead must move down too far, which results in a too large cutting angle, and the stylus literally digs into the record!
If the installation height is too low, then the stylus rests too early on the record and the cutting angle is less than 90°.
In addition, the contact point must be exactly in line with the center of the record.
So that nothing goes wrong, I have mounted a cross laser. This is set to plate height and plate center. With the laser, the adjustment of the needle is then quite simple.
The short video shows the laser on the stylus
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
You do not have the required permissions to view the files attached to this post.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
2022_06 First attempt at cutting - video
I have made a video of the complete system in the attachment. It was one of the first cutting attempts with the new cutterhead.
The suction worked well and the cut thread was extracted well.
The depth of cut was fine right off the bat (coincidentally).
I did the first cut with lower amp power for fear of killing a speaker right away.
The short video shows the setup at work
ATTENTION! turn down the headphones, the exhaust is very loud (I didn't have time to build a muffler housing yet)
I have made a video of the complete system in the attachment. It was one of the first cutting attempts with the new cutterhead.
The suction worked well and the cut thread was extracted well.
The depth of cut was fine right off the bat (coincidentally).
I did the first cut with lower amp power for fear of killing a speaker right away.
The short video shows the setup at work
ATTENTION! turn down the headphones, the exhaust is very loud (I didn't have time to build a muffler housing yet)
You do not have the required permissions to view the files attached to this post.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
2022_06 First cutting test - test setup
Test with White Noise:
I saved the "white noise" signal to a player - in my case an Ipod mp3. It doesn't provide the best signal in this form, but it's good enough compared to my cutterhead. - Once my cutter cuts as well as the mp3, I use signals from the sound card. But at the moment this is too much work for me spatially. You know, I work in the kitchen and living room! My wife will certainly not play along with me laying mountains of cables across the room - and I want to keep her longer.
The "white noise" signal was not processed (NO IRIAA)! I sent it directly into the amplifier - no filter, no anything. The reason is the excess of high frequencies. If I add an IRIAA to that as well, I'll surely fry my head.
After that it is cut and then played back on the turntable. From the turntable, the signal then travels via the amplifier (with built-in RIAA) to my PC. There I record the signal then with the software (Cakewalk) and provide it with the PlugIn "IRIAA". Thus again a clear "White Noise" signal should be represented as straight characteristic curve.
I do the signal analysis with "VA-Visual Analyzer 2021". This then gives an FFT analysis.
It took me a few days to find the necessary software packages and to understand how they work, but now it works. (Unfortunately my good PC sound card died, but I quickly bought a new one).
2022_06 First cutting attempt - First results
1st test with White Noise - Stereo:
As you can see, it's not a straight line! But that would be too easy!
a.) From other contributions I already know this mountain in the middle (in my case at about 600Hz). I have no experience with such signals yet, but I think the mountain should be rather at 1kHz or higher?! What could be the reason for that?
b.) This smaller mountain at approx. 12kHz is also familiar to me from other characteristic curves.
c.) I find the signal peaks at 10kHz bad. There must be resonances involved - but which ones?
d.) I find the frequency peaks at 1800Hz very bad. They are not nice at all and first negative, then positive - and present in both loudspeakers, even if a bit less in the right channel. They don't belong there at all.
In comparison, the measurement of the original "White Noise". This is how it should look!
2022_06 Further test cuts
2nd test with White Noise - this time only left, then only right channel:
Of course there is still no straight line! The other problems (see stereo cut) have also remained the same.
What can be seen, however, is the recognizable channel separation, even if only in the frequency range 600Hz to 8kHz.
The fact that a signal is still cut on the other channel is due to something.
Aber woran liegt es?
On the V-spring?
The geometry of the torque tube?
At the retaining wire of the torque tube?
An unwanted mechanical transmission from one channel to the other?
A loose connection between two components?
Too soft or too hard speaker suspension?
The following is striking:
a. The bad signal peaks at 10kHz (resonances ?) are only in the driven channel - in the non-driven channel this interference peak is (almost) not present.
b. The "high mountain" was at about 600Hz for the stereo signal. In single channel operation it drops to 400 - 500Hz.
c. The signal peak in the 12kHz range is always there - even in single channel operation.
PICTURE ABOVE: ONLY THE LEFT CHANNEL
PICTURE BELOW: ONLY THE RIGHT CHANNEL
2022_06 More test cuts
3rd test with white noise - stereo:
Now a systematic approach is required. So the first thing I did was to remove the holding wire of the torsion bar from its attachment - because that is the easiest way. If nothing changes, then I can at least rule out the thickness of the wire (too soft or too stiff), and I can also tick off any resonances that might come from the wire.
The result: everything is the same as before.
Test with White Noise:
I saved the "white noise" signal to a player - in my case an Ipod mp3. It doesn't provide the best signal in this form, but it's good enough compared to my cutterhead. - Once my cutter cuts as well as the mp3, I use signals from the sound card. But at the moment this is too much work for me spatially. You know, I work in the kitchen and living room! My wife will certainly not play along with me laying mountains of cables across the room - and I want to keep her longer.
The "white noise" signal was not processed (NO IRIAA)! I sent it directly into the amplifier - no filter, no anything. The reason is the excess of high frequencies. If I add an IRIAA to that as well, I'll surely fry my head.
After that it is cut and then played back on the turntable. From the turntable, the signal then travels via the amplifier (with built-in RIAA) to my PC. There I record the signal then with the software (Cakewalk) and provide it with the PlugIn "IRIAA". Thus again a clear "White Noise" signal should be represented as straight characteristic curve.
I do the signal analysis with "VA-Visual Analyzer 2021". This then gives an FFT analysis.
It took me a few days to find the necessary software packages and to understand how they work, but now it works. (Unfortunately my good PC sound card died, but I quickly bought a new one).
2022_06 First cutting attempt - First results
1st test with White Noise - Stereo:
As you can see, it's not a straight line! But that would be too easy!
a.) From other contributions I already know this mountain in the middle (in my case at about 600Hz). I have no experience with such signals yet, but I think the mountain should be rather at 1kHz or higher?! What could be the reason for that?
b.) This smaller mountain at approx. 12kHz is also familiar to me from other characteristic curves.
c.) I find the signal peaks at 10kHz bad. There must be resonances involved - but which ones?
d.) I find the frequency peaks at 1800Hz very bad. They are not nice at all and first negative, then positive - and present in both loudspeakers, even if a bit less in the right channel. They don't belong there at all.
In comparison, the measurement of the original "White Noise". This is how it should look!
2022_06 Further test cuts
2nd test with White Noise - this time only left, then only right channel:
Of course there is still no straight line! The other problems (see stereo cut) have also remained the same.
What can be seen, however, is the recognizable channel separation, even if only in the frequency range 600Hz to 8kHz.
The fact that a signal is still cut on the other channel is due to something.
Aber woran liegt es?
On the V-spring?
The geometry of the torque tube?
At the retaining wire of the torque tube?
An unwanted mechanical transmission from one channel to the other?
A loose connection between two components?
Too soft or too hard speaker suspension?
The following is striking:
a. The bad signal peaks at 10kHz (resonances ?) are only in the driven channel - in the non-driven channel this interference peak is (almost) not present.
b. The "high mountain" was at about 600Hz for the stereo signal. In single channel operation it drops to 400 - 500Hz.
c. The signal peak in the 12kHz range is always there - even in single channel operation.
PICTURE ABOVE: ONLY THE LEFT CHANNEL
PICTURE BELOW: ONLY THE RIGHT CHANNEL
2022_06 More test cuts
3rd test with white noise - stereo:
Now a systematic approach is required. So the first thing I did was to remove the holding wire of the torsion bar from its attachment - because that is the easiest way. If nothing changes, then I can at least rule out the thickness of the wire (too soft or too stiff), and I can also tick off any resonances that might come from the wire.
The result: everything is the same as before.
You do not have the required permissions to view the files attached to this post.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
2022_06 More test cuts
4th test with music and equalizer - stereo:
Out of pure curiosity and the inner desire to also cut music once, I did the following. This software from "Cakewalk" also has a Sonitus fx equalizer built in. I thought if I invert the cut signal from the "White Noise", then I get an approximate line of the necessary EQ:
This is roughly how I drew the first EQ curve: Then I processed the piece of music with the equalizer and an IRIAA and transferred it to the I-Pod - and cut it.
What the heck, what could go wrong?
Here is the first music audio.
[/b]
After that I noticed that I had not yet reattached the wire from the torque tube. I then quickly did that again and immediately again (with a little more power taken).
[/b]
The pros notice two things right away. First, the recording speed is too slow, resulting in too fast a playback speed. This was due to a small error in my equipment, which I noticed but couldn't solve right away. In the meantime I solved it (two weeks later, because I had to disassemble my whole turntable drive unit to get to the print plate - small mistake - big work).
And second - it doesn't really sound good. I'm not satisfied - but at least I had a test run, which was by far better than what I had before.
4th test with music and equalizer - stereo:
Out of pure curiosity and the inner desire to also cut music once, I did the following. This software from "Cakewalk" also has a Sonitus fx equalizer built in. I thought if I invert the cut signal from the "White Noise", then I get an approximate line of the necessary EQ:
This is roughly how I drew the first EQ curve: Then I processed the piece of music with the equalizer and an IRIAA and transferred it to the I-Pod - and cut it.
What the heck, what could go wrong?
Here is the first music audio.
[/b]
After that I noticed that I had not yet reattached the wire from the torque tube. I then quickly did that again and immediately again (with a little more power taken).
[/b]
The pros notice two things right away. First, the recording speed is too slow, resulting in too fast a playback speed. This was due to a small error in my equipment, which I noticed but couldn't solve right away. In the meantime I solved it (two weeks later, because I had to disassemble my whole turntable drive unit to get to the print plate - small mistake - big work).
And second - it doesn't really sound good. I'm not satisfied - but at least I had a test run, which was by far better than what I had before.
You do not have the required permissions to view the files attached to this post.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
2022_06 More test cuts
Next are some test cuts (white noise) with different recording volumes. Unfortunately, my amplifier has only 2x120W at 4Ohm - I'm already looking for something stronger, but actually do not want such a PA beater, but better something with 2x300W (4Ohm) Class AB.
But no matter now. I'll push my amp and make five recordings. I want to see if there is a difference in the characteristics or resonances for quiet or loud recordings.
The respective dB readings refer to my VU meter readings (-5dB; -3dB; -1dB; 0dB; +1dB). At +1dB, the speakers on the Cutterhead are already very loud! (My first cutterhead already died at -3dB during this test).
In principle, the signals are all the same. In detail, it can be determined that the louder the recording was made, the further the frequency in the "big mountain" dropped.
If at quiet recording (-5dB) the maximum was at ~650Hz, at +1dB this frequency dropped to 500Hz.
The same behavior can be noticed at the frequency peak ~8.5kHz. This drops to 7.5kHz.
Likewise at 12kHz (although only slightly decreasing).
At the frequency peak in the 1800Hz range and those at 10kHz, this behavior is not noticeable.
I already feel this way permanently. One answer and ten new questions!
I will build new speakers. - Just to determine if the spiders (which center the coils) are the problem.
See you soon, best regards from Austria
Thomas
Next are some test cuts (white noise) with different recording volumes. Unfortunately, my amplifier has only 2x120W at 4Ohm - I'm already looking for something stronger, but actually do not want such a PA beater, but better something with 2x300W (4Ohm) Class AB.
But no matter now. I'll push my amp and make five recordings. I want to see if there is a difference in the characteristics or resonances for quiet or loud recordings.
The respective dB readings refer to my VU meter readings (-5dB; -3dB; -1dB; 0dB; +1dB). At +1dB, the speakers on the Cutterhead are already very loud! (My first cutterhead already died at -3dB during this test).
In principle, the signals are all the same. In detail, it can be determined that the louder the recording was made, the further the frequency in the "big mountain" dropped.
If at quiet recording (-5dB) the maximum was at ~650Hz, at +1dB this frequency dropped to 500Hz.
The same behavior can be noticed at the frequency peak ~8.5kHz. This drops to 7.5kHz.
Likewise at 12kHz (although only slightly decreasing).
At the frequency peak in the 1800Hz range and those at 10kHz, this behavior is not noticeable.
I already feel this way permanently. One answer and ten new questions!
I will build new speakers. - Just to determine if the spiders (which center the coils) are the problem.
See you soon, best regards from Austria
Thomas
You do not have the required permissions to view the files attached to this post.
Re: The farmer and his plow - When I set out to build a record cutter
Hi,
Looks like you are making progress on your head design.
Your main system resonance arises from two parameters - moving mass and stiffness. Generally, you would like to get the system resonance up as high as possible, but there are some tradeoffs. If you make the head stiffer to raise the resonance, it will require more power to drive the head in the low frequency range (below resonance). A stiffer head will be less influenced by the mechanical resistance to cutting. Just like a stiff (low output impedance) unregulated power supply is not affected so much by an external load. However, if you make it too stiff, you will never be able to fully modulate the grooves before you run out of amp power or driver power handling. This is dependent on the BL factor of you drivers. A higher BL generates more force for a given drive current in the coils and will work against the stiffness below resonance and produces acceleration in the mass controlled region above resonance. Next up is the moving mass. If your moving mass is too high it will lower your resonance and work against you at high frequencies (F=MA and F = BLi). Generally, you want to minimize moving mass since it has such a negative affect on head performance. Note that to double the resonance frequency, you need to increase the stiffness by 4X or reduce the mass by 1/4 or some combination of the two.
Welcome to the real world on the secondary resonance front. I find this to be the hardest thing to analyze and correct (I'm not a mechanical engineer). There all kinds of hidden spring mass systems lurking be it push rods, cones, springs, voice coil forms, stylus shanks, etc. Killing them is not easy and will take you some trial and error. Now you can see why I indicated that getting a clean responding open loop head design is so important before you try to close a servo loop. That would be a tough job given the gain/phase response you are showing.
That said, I think you initial results look quite good and you should be able to use this head as is with some corrective EQ.
All in all your skills and build is quite amazing. Keep up the great work!
Mark
Looks like you are making progress on your head design.
Your main system resonance arises from two parameters - moving mass and stiffness. Generally, you would like to get the system resonance up as high as possible, but there are some tradeoffs. If you make the head stiffer to raise the resonance, it will require more power to drive the head in the low frequency range (below resonance). A stiffer head will be less influenced by the mechanical resistance to cutting. Just like a stiff (low output impedance) unregulated power supply is not affected so much by an external load. However, if you make it too stiff, you will never be able to fully modulate the grooves before you run out of amp power or driver power handling. This is dependent on the BL factor of you drivers. A higher BL generates more force for a given drive current in the coils and will work against the stiffness below resonance and produces acceleration in the mass controlled region above resonance. Next up is the moving mass. If your moving mass is too high it will lower your resonance and work against you at high frequencies (F=MA and F = BLi). Generally, you want to minimize moving mass since it has such a negative affect on head performance. Note that to double the resonance frequency, you need to increase the stiffness by 4X or reduce the mass by 1/4 or some combination of the two.
Welcome to the real world on the secondary resonance front. I find this to be the hardest thing to analyze and correct (I'm not a mechanical engineer). There all kinds of hidden spring mass systems lurking be it push rods, cones, springs, voice coil forms, stylus shanks, etc. Killing them is not easy and will take you some trial and error. Now you can see why I indicated that getting a clean responding open loop head design is so important before you try to close a servo loop. That would be a tough job given the gain/phase response you are showing.
That said, I think you initial results look quite good and you should be able to use this head as is with some corrective EQ.
All in all your skills and build is quite amazing. Keep up the great work!
Mark
Re: The farmer and his plow - When I set out to build a record cutter
hello
i wrote thomas directly in german. but i think just for all the followers some short hints.
i just can repeat and repeat every few years. these soft spider coils will never work....
i posted that almost in the beginning of lathetrolls. and there are some links
that covers some of the basic topics for a good and nice cutting system.
if you have all that. its a very good start. and way better than certain commercial available solutions. or at least the stuff can upgrade it.
https://www.floka.com/lofi/lofi_cutting.html
i t was done 21 years ago.and a total failure.
https://www.floka.com/lofi/lofihead_due.html.
but using good hifi tweeters bring you quite far.
i used gradient back in the time was not so bad.
https://www.floka.com/lofi/lofihead_uno.html
and yes. mark's said it perfectly. if you dont have good and stable open loop head. dont even think about feedback.
use the coils to see what happens. closing the FB loop means pushing the self destruction button in most cases
i just repost in hope that not everybody does the same experiments over and over and gets frustrated
there are many more important topics to solve.
styli polishing, blank experiments ,
dsp forward compensated lofi heads.
diy galvanics
diy backjard pressing
alternative materials
...
...
the list is long
happy cutting.
i wrote thomas directly in german. but i think just for all the followers some short hints.
i just can repeat and repeat every few years. these soft spider coils will never work....
i posted that almost in the beginning of lathetrolls. and there are some links
that covers some of the basic topics for a good and nice cutting system.
if you have all that. its a very good start. and way better than certain commercial available solutions. or at least the stuff can upgrade it.
https://www.floka.com/lofi/lofi_cutting.html
i t was done 21 years ago.and a total failure.
https://www.floka.com/lofi/lofihead_due.html.
but using good hifi tweeters bring you quite far.
i used gradient back in the time was not so bad.
https://www.floka.com/lofi/lofihead_uno.html
and yes. mark's said it perfectly. if you dont have good and stable open loop head. dont even think about feedback.
use the coils to see what happens. closing the FB loop means pushing the self destruction button in most cases
i just repost in hope that not everybody does the same experiments over and over and gets frustrated
there are many more important topics to solve.
styli polishing, blank experiments ,
dsp forward compensated lofi heads.
diy galvanics
diy backjard pressing
alternative materials
...
...
the list is long
happy cutting.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
Unfortunately I was on the road the last days and could not report further. As you have surely noticed, I am not yet at my current state but in my reports still a few weeks behind; but I'm catching up....
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
Hi Mark,markrob wrote: ↑Thu Jul 07, 2022 8:07 pmHi,
Looks like you are making progress on your head design.
Your main system resonance arises from two parameters - moving mass and stiffness. Generally, you would like to get the system resonance up as high as possible, but there are some tradeoffs. I...
Mark
thanks for this post - it always helps to benefit from the experience of professionals.
The relationships between mass and stiffness are very well explained. I was aware that the moving mass should be as low as possible so that the coils can still move the stylus at high frequencies. However, my thought error was that I thought a "softer" suspension would make cutting easier (in all frequency ranges) without considering that I would get too low a main resonance.
To get experience I replaced the "spider coils" with metal suspensions in June.
I have to increase the stiffness further. For reducing the mass I still have some ideas.
Your hint to design a cutting head first before I worry about feedback coils I have understood and am about to implement it that way. I also have a few approaches for finding and killing the secondary resonances. - Whether I will succeed in this? I will find out.
Are carbon cone and drive rod the wrong choice?
Kind regards
Thomas
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
Hello Flo,flozki wrote: ↑Fri Jul 08, 2022 5:14 amhello
i wrote thomas directly in german. but i think just for all the followers some short hints.
i just can repeat and repeat every few years. these soft spider coils will never work....
i posted that almost in the beginning of lathetrolls. and there are some links
that covers some of the basic topics for a good and nice cutting system. ...
happy cutting.
Thank you for your competent advice (specialists do more). That spider coils will never work I gladly accept as a hint from an experienced person. I wanted to make my own experiences with it. In my following report I show the differences between spider coils and metal coils at the same loudspeaker - the results could be misleading and therefore I thank you already now for your hints.
I have already ordered tweeters (I still want to present my experience with the existing speakers).
For reassurance, I can share that I like to experiment and have been a little depressed after each failed attempt, but am far from frustrated (at this time).
Due to my limited capabilities, I am currently unable to work on "styli polishing" and "blank experiments" for me are limited to anti-static, polish and temperature.
What do you mean by "diy galvanics"? - galvanize the used metal parts?
Kind regards
Thomas
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
2022_06 The polarity of the speakers
When analyzing the signal of purchased records, one can see that the polarity of the speakers are obviously reversed. So this was done when cutting the records, probably so that the movement of the cutting graver can be done "more easily". So left channel and right channel do not work simultaneously on the stereo signal and push the graver down at the same time. One channel is opposite polarity, and so the two speakers "help" each other cut.
So I recorded a "white noise" signal twice. First with speakers of the same polarity and once I reversed the poles on the left channel.
The comparison picture shows the difference:
The main resonance drops from 600Hz to 350Hz! (It gets worse and worse!)
When analyzing the signal of purchased records, one can see that the polarity of the speakers are obviously reversed. So this was done when cutting the records, probably so that the movement of the cutting graver can be done "more easily". So left channel and right channel do not work simultaneously on the stereo signal and push the graver down at the same time. One channel is opposite polarity, and so the two speakers "help" each other cut.
So I recorded a "white noise" signal twice. First with speakers of the same polarity and once I reversed the poles on the left channel.
The comparison picture shows the difference:
The main resonance drops from 600Hz to 350Hz! (It gets worse and worse!)
You do not have the required permissions to view the files attached to this post.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
2022_06 Loudspeaker with metal ring coil suspension instead of spider
To determine the influence of the coil suspension on the main resonance (and on the secondary resonances), I decided to change the coil suspension from "spider" to "metal ring".
To do this, I carefully ground down the original suspension to make it a little thinner (because I had the impression that the original was too stiff). Additionally I reduced the weight a bit by drilling holes. More consistent material removal or a redesign could save even more weight - but it's a first attempt:
The speakers in comparison:
The installation of the new speakers:
To determine the influence of the coil suspension on the main resonance (and on the secondary resonances), I decided to change the coil suspension from "spider" to "metal ring".
To do this, I carefully ground down the original suspension to make it a little thinner (because I had the impression that the original was too stiff). Additionally I reduced the weight a bit by drilling holes. More consistent material removal or a redesign could save even more weight - but it's a first attempt:
The speakers in comparison:
The installation of the new speakers:
You do not have the required permissions to view the files attached to this post.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
2022_06 First tests with the loudspeaker with coil suspension made of metal ring instead of spider
The polarity on the left speaker remains reversed for all subsequent tests.
The first test with "White Noise" signal shows that the main resonance has dropped even further and is now at 280Hz.
On superficial inspection, one might think that the Spider suspension is better. That would contradict Flo's statements. But when you consider that the weight of the metal ring suspension is much higher, it put the first impression into perspective. Mark had explained very well how badly high weight affects the result.
In addition, I must point out that I have already ground the metal rings very thin, so that in the end the coil sits almost as "soft" as the spider version. Thanks to the experience of Flo, I will not rely on the Spider!
Another test is to show the stereo effect of the Cutterhead.
You can see the different signal strength and the main resonance at 350Hz (instead of 280Hz in stereo mode).
RIGHT CHANNEL: LEFT CHANNEL:
Attached are the five recordings of "White Noise" in comparison
Cuttet (unmodified) "White Noise" with subsequent IRIAA
Channel Right and Left "normally" poled (coil with spider) Channel Right "normal" polarity and Left "reverse" polarity (coil with spider) Channel Right "normal" polarity and Left "reverse" polarity (coil with metal ring) Channel Right "normal" polarity and Left "reverse" polarity (coil with metal ring) RIGHT channel only Channel Right "normal" polarity and Left "reverse" polarity (coil with metal ring) LEFT channel only[/b]
The polarity on the left speaker remains reversed for all subsequent tests.
The first test with "White Noise" signal shows that the main resonance has dropped even further and is now at 280Hz.
On superficial inspection, one might think that the Spider suspension is better. That would contradict Flo's statements. But when you consider that the weight of the metal ring suspension is much higher, it put the first impression into perspective. Mark had explained very well how badly high weight affects the result.
In addition, I must point out that I have already ground the metal rings very thin, so that in the end the coil sits almost as "soft" as the spider version. Thanks to the experience of Flo, I will not rely on the Spider!
Another test is to show the stereo effect of the Cutterhead.
You can see the different signal strength and the main resonance at 350Hz (instead of 280Hz in stereo mode).
RIGHT CHANNEL: LEFT CHANNEL:
Attached are the five recordings of "White Noise" in comparison
Cuttet (unmodified) "White Noise" with subsequent IRIAA
Channel Right and Left "normally" poled (coil with spider) Channel Right "normal" polarity and Left "reverse" polarity (coil with spider) Channel Right "normal" polarity and Left "reverse" polarity (coil with metal ring) Channel Right "normal" polarity and Left "reverse" polarity (coil with metal ring) RIGHT channel only Channel Right "normal" polarity and Left "reverse" polarity (coil with metal ring) LEFT channel only[/b]
You do not have the required permissions to view the files attached to this post.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
2022_06 The fuse circuit (part 2)
Out of curiosity, I recorded a piece of music with the same settings. It sounded bad! It was quiet, no bass and no treble. Since my amplifier is not powerful (2x120W) but was already at near full power output, I had the following thought: my built-in RC speaker fuse circuit could perhaps be affecting the frequency response and causing losses.
So I shorted the RC circuit on one channel and measured with "white noise" signal (I left the fuses installed and replaced them with 250mA fuses): The frequency changes and losses are clearly visible.
Out of curiosity, I recorded a piece of music with the same settings. It sounded bad! It was quiet, no bass and no treble. Since my amplifier is not powerful (2x120W) but was already at near full power output, I had the following thought: my built-in RC speaker fuse circuit could perhaps be affecting the frequency response and causing losses.
So I shorted the RC circuit on one channel and measured with "white noise" signal (I left the fuses installed and replaced them with 250mA fuses): The frequency changes and losses are clearly visible.
You do not have the required permissions to view the files attached to this post.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
2022_07 White Noise Signal (without RC speaker safety circuit)
So I decided to short the RC circuit on both channels.
The signals were louder after that and the main resonance increased to 400Hz.
So I decided to short the RC circuit on both channels.
The signals were louder after that and the main resonance increased to 400Hz.
You do not have the required permissions to view the files attached to this post.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
2022_07 Small mistake - big impact: Do witches exist?
During another test, I wanted to cut even louder. But then something very strange happened. I thought I was recording an echo!
Listen to this!
[/b]
The recording was suddenly much quieter. But when I played only one channel (left only or right only) the speakers on the cutterhead were very loud again!
While troubleshooting I noticed that both fuses were melted! Still sound and music came to the cutterhead! Are there witches?
So I searched further. From the amplifier, the signals went directly to the circuit board with the fuses and the now shorted RC circuit (which was already working fine). And from the circuit board directly to the speakers.
The mistake was a stupid rookie mistake - I had put the fuses on the negative pole instead of the positive pole. In principle it wouldn't matter - but not in my case. Because I still had the VU meters connected! The control of the VU meters runs over a circuit board, where both minus poles are connected. Obviously the speakers got ground via the VU meters.
The error was quickly fixed, both fuses were replaced and the speakers remained intact.
During another test, I wanted to cut even louder. But then something very strange happened. I thought I was recording an echo!
Listen to this!
[/b]
The recording was suddenly much quieter. But when I played only one channel (left only or right only) the speakers on the cutterhead were very loud again!
While troubleshooting I noticed that both fuses were melted! Still sound and music came to the cutterhead! Are there witches?
So I searched further. From the amplifier, the signals went directly to the circuit board with the fuses and the now shorted RC circuit (which was already working fine). And from the circuit board directly to the speakers.
The mistake was a stupid rookie mistake - I had put the fuses on the negative pole instead of the positive pole. In principle it wouldn't matter - but not in my case. Because I still had the VU meters connected! The control of the VU meters runs over a circuit board, where both minus poles are connected. Obviously the speakers got ground via the VU meters.
The error was quickly fixed, both fuses were replaced and the speakers remained intact.
You do not have the required permissions to view the files attached to this post.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
2022_07 New test cuts with music
I adjusted the equalizer a bit and made first test cuts with music.
In the meantime, the results sound better, although the measurement results are not yet good.
Here are a few samples: (In the last example, a cut thread formed briefly that was not immediately sucked out. You can hear how the thread pulls and tugs on the needle. I had to give the turntable a little push at this point to make it jump over the groove...).
I also like the stereo effect.
[/b]
In the meantime I have ordered new tweeters, because I think that I will not achieve the desired results with this speaker constellation. When these are delivered I can continue building.
See you soon, best regards from Austria
Thomas
I adjusted the equalizer a bit and made first test cuts with music.
In the meantime, the results sound better, although the measurement results are not yet good.
Here are a few samples: (In the last example, a cut thread formed briefly that was not immediately sucked out. You can hear how the thread pulls and tugs on the needle. I had to give the turntable a little push at this point to make it jump over the groove...).
I also like the stereo effect.
[/b]
In the meantime I have ordered new tweeters, because I think that I will not achieve the desired results with this speaker constellation. When these are delivered I can continue building.
See you soon, best regards from Austria
Thomas
You do not have the required permissions to view the files attached to this post.
-
ilovedrums247
- Posts: 35
- Joined: Mon Apr 02, 2018 10:28 am
- Contact:
Re: The farmer and his plow - When I set out to build a record cutter
What beautiful machining!
It's very inspiring work,and test cuts sound really good.
thank you for sharing.
It's very inspiring work,and test cuts sound really good.
thank you for sharing.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
2022_07 In search of resonances
First of all, I would like to thank you for your positive feedbacks. These feedbacks motivate me to continue reporting.
However, I look forward to any positive, but also any negative criticism (constructive), because I have a tendency to get lost in my analyses. So if you notice an analysis that I misinterpret, please please please bring me back on the path of objective perception.
Since the new speakers have not yet arrived, I'm going to look for the resonant frequencies in the existing cutterhead to gain experience.In the process, I want to try to make changes and measure the reactions to them. Since I still have quite little experience, this will be an interesting journey for me.
The first change I make is to the V-spring. For this I attach two neodymium magnets on each side. These should limit the "freedom of movement" of the spring somewhat. This experiment should bring about a change in any case, because otherwise it would not matter how the V-spring looks like?
What do I expect from this procedure?:
First, it restricts the freedom of movement of the V-spring. Therefore, it acts like an additional weight that must be moved by the speakers. Consequently, the main resonant frequency should decrease.Second, I can tell if the resonant frequency is in the 1.5kHz range or at 10-12kHz from the V-spring.
Images: This is how the magnets were attached to the V-spring:
(without magnets) - (with vmagnets)
With these changes, "white noise" test cuts were made again - at three different volumes.
Since I cut with higher volume, I had to increase the distance between the grooves to 300µm (0.0118 inch). (The volume refers to the power meter of my amplifier).
Here is the groove pattern:
The evaluation of the frequency band is also interesting for me.
The louder the recording, the lower the faulty resonances at 1.5 kHz and at 10 kHz. However, I have to relativize the result again: With the (very) loud recording there were already distortions. These could also have influenced the resonance curve.
Comparison with and without V-spring damping:
Comparing the signals without magnets on the V-spring, the following can be seen:
First, the main resonance drops (slightly) from 290Hz to 280Hz (as expected).
Second, the resonance at 1.2kHz remains unaffected.
Third, the resonance at 10kHz appears a bit more damped.
Signal comparison with and without magnets on the V-spring:
Increase the main resonance:
As a next step I want to shift the main resonance more towards 1kHz. There are two ways to do this (as markrob has already described): Increase the stiffness of the head and/or reduce the weight of the moving masses.
In my case, that means making the speakers much lighter with the metal ring. The metal ring (with its aluminum washer) is drilled to reduce the weight from 7.6g to 5.4g, but that's still about 4 grams too much! For this I need new speakers (ordered) to rebuild the suspension to get you down to about 1g to 2g. (I don't want to and can't separate the existing speakers from the coil without destroying them).
Or:
I'll use the speakers with the spider suspension again. Even at the risk of Flozki stoning me, because he has explained quite clearly that with the spider coils no satisfactory result is to be achieved. But since I already have them, I can do some tests with them until the ordered tweeters arrive.
The advantage of my Spider-Coils is that they already have a very low weight, which I can hardly reduce further (with my possibilities). But I can increase the stiffness of the head, which should result in an increased frequency of the main resonance.
To do this, I use my good friend "Mr. Epoxy Resin."
So I paint the spiders with epoxy resin. This should dramatically increase the stiffness.
Additionally, I coat one of the two speaker cones including the drive rod with epoxy resin as well - to increase the stiffness here as well. This will make this channel a bit heavier (lowering the main resonance), but it might also result in a reduction of the high frequency resonances. So I can also test this effect right away as well.
Loudspeaker - Spider with epoxy resin:
For the left channel, I only epoxied the spider diaphragm.
On the right channel, I epoxied the spider diaphragm, the cone and the drive rod.
The third picture shows both speakers before installation
After 24 hours of curing, I can already see that the diaphragm is much firmer than before when I "finger test" it by pressing on it.
What do I expect from this intervention?
First, an increase in the frequency of the main resonance.
Secondly, a recognizable difference between the two channels. The findings from this should influence the further tests.
I myself am already very curious about the measurement results.
See you soon, best regards from Austria
Thomas
First of all, I would like to thank you for your positive feedbacks. These feedbacks motivate me to continue reporting.
However, I look forward to any positive, but also any negative criticism (constructive), because I have a tendency to get lost in my analyses. So if you notice an analysis that I misinterpret, please please please bring me back on the path of objective perception.
Since the new speakers have not yet arrived, I'm going to look for the resonant frequencies in the existing cutterhead to gain experience.In the process, I want to try to make changes and measure the reactions to them. Since I still have quite little experience, this will be an interesting journey for me.
The first change I make is to the V-spring. For this I attach two neodymium magnets on each side. These should limit the "freedom of movement" of the spring somewhat. This experiment should bring about a change in any case, because otherwise it would not matter how the V-spring looks like?
What do I expect from this procedure?:
First, it restricts the freedom of movement of the V-spring. Therefore, it acts like an additional weight that must be moved by the speakers. Consequently, the main resonant frequency should decrease.Second, I can tell if the resonant frequency is in the 1.5kHz range or at 10-12kHz from the V-spring.
Images: This is how the magnets were attached to the V-spring:
(without magnets) - (with vmagnets)
With these changes, "white noise" test cuts were made again - at three different volumes.
Since I cut with higher volume, I had to increase the distance between the grooves to 300µm (0.0118 inch). (The volume refers to the power meter of my amplifier).
Here is the groove pattern:
The evaluation of the frequency band is also interesting for me.
The louder the recording, the lower the faulty resonances at 1.5 kHz and at 10 kHz. However, I have to relativize the result again: With the (very) loud recording there were already distortions. These could also have influenced the resonance curve.
Comparison with and without V-spring damping:
Comparing the signals without magnets on the V-spring, the following can be seen:
First, the main resonance drops (slightly) from 290Hz to 280Hz (as expected).
Second, the resonance at 1.2kHz remains unaffected.
Third, the resonance at 10kHz appears a bit more damped.
Signal comparison with and without magnets on the V-spring:
Increase the main resonance:
As a next step I want to shift the main resonance more towards 1kHz. There are two ways to do this (as markrob has already described): Increase the stiffness of the head and/or reduce the weight of the moving masses.
In my case, that means making the speakers much lighter with the metal ring. The metal ring (with its aluminum washer) is drilled to reduce the weight from 7.6g to 5.4g, but that's still about 4 grams too much! For this I need new speakers (ordered) to rebuild the suspension to get you down to about 1g to 2g. (I don't want to and can't separate the existing speakers from the coil without destroying them).
Or:
I'll use the speakers with the spider suspension again. Even at the risk of Flozki stoning me, because he has explained quite clearly that with the spider coils no satisfactory result is to be achieved. But since I already have them, I can do some tests with them until the ordered tweeters arrive.
The advantage of my Spider-Coils is that they already have a very low weight, which I can hardly reduce further (with my possibilities). But I can increase the stiffness of the head, which should result in an increased frequency of the main resonance.
To do this, I use my good friend "Mr. Epoxy Resin."
So I paint the spiders with epoxy resin. This should dramatically increase the stiffness.
Additionally, I coat one of the two speaker cones including the drive rod with epoxy resin as well - to increase the stiffness here as well. This will make this channel a bit heavier (lowering the main resonance), but it might also result in a reduction of the high frequency resonances. So I can also test this effect right away as well.
Loudspeaker - Spider with epoxy resin:
For the left channel, I only epoxied the spider diaphragm.
On the right channel, I epoxied the spider diaphragm, the cone and the drive rod.
The third picture shows both speakers before installation
After 24 hours of curing, I can already see that the diaphragm is much firmer than before when I "finger test" it by pressing on it.
What do I expect from this intervention?
First, an increase in the frequency of the main resonance.
Secondly, a recognizable difference between the two channels. The findings from this should influence the further tests.
I myself am already very curious about the measurement results.
See you soon, best regards from Austria
Thomas
You do not have the required permissions to view the files attached to this post.
-
farmersplow
- Posts: 376
- Joined: Thu Jun 17, 2021 3:43 am
- Location: Austria - Vienna
Re: The farmer and his plow - When I set out to build a record cutter
First tests with the speakers - spider with epoxy resin:
After I install the speakers in my cutterhead, I will do "white noise" signal testing.
Magnets were removed from the V-spring!!!
Speakers with SPIN were inserted.
Spiders hardened with epoxy.
I am positively amazed at the result! Here in comparison with the same speakers without epoxy resin impregnated coil spiders:
The main resonance frequency has increased from 580Hz to about 820Hz!
The "secondary resonance" at 1.8kHz has remained about the same.
The resonances (~9kHz) have been weakened somewhat.
The resonance at about 12kHz has become more intense, and dropped to about 11kHz.
The right channel (RED: speaker with the epoxy coated cone and drive rod has also reached a higher frequency than the left (Green) channel due to the increased stiffness.
At the same time, the values of the right channel in the range below 820Hz also have -3dB less power.
So far, the expectations have been met.
Unclear to me is the Woppel (blue circle) at 300Hz???
Does anyone have an explanation?
Here also the comparison:
Loudspeakers with epoxy resin impregnated coil spiders vs loudspeakers with metal ring
The main resonance frequency has increased from 280Hz to about 820Hz!
The "secondary resonance" has increased from 1.2kHz to 1.8kHz.
The resonances (~9kHz) have been somewhat attenuated.
The resonance at about 10kHz has become more intense, and increased to about 11kHz.
The high frequency range from 12kHz has "calmed down"
I expect, when I receive the new resonant speakers (with harder metal ring), to see a similar improvement in the main resonance with these metal ring versions.
Further search for the solutions for resonance frequencies
To find out if the drive rods cause the resonances in the 12kHz range, I disassembled the speakers again and mounted small dampers on the drive rods.
These dampers are made of silicone and have a weight of about 0.5g per speaker (2 pieces per speaker).
What influence will this intervention have on the frequency image?
What do I expect from this intervention?:
First, a small decrease in the frequency of the main resonance because the coils now have to move more weight.
Second, some reduction in resonances in the higher frequency range (above 6kHz to 15kHz).
I suspect that resonances in the higher frequency range (above 6kHz to 15kHz), can only come from small thin components. I further suspect that my brass bronze rod, 5mm long, has a much higher resonant frequency than the carbon cones, which are softer and larger. These cones are more likely to be below the 10kHz (perhaps below 3kHz, causing the spurious resonance at 1.8kHz?). I have no idea about the carbon rods, which are 25mm long and 3mm in diameter.
I am very curious about the measurement results myself.
See you soon, best regards from Austria
Thomas
After I install the speakers in my cutterhead, I will do "white noise" signal testing.
Magnets were removed from the V-spring!!!
Speakers with SPIN were inserted.
Spiders hardened with epoxy.
I am positively amazed at the result! Here in comparison with the same speakers without epoxy resin impregnated coil spiders:
The main resonance frequency has increased from 580Hz to about 820Hz!
The "secondary resonance" at 1.8kHz has remained about the same.
The resonances (~9kHz) have been weakened somewhat.
The resonance at about 12kHz has become more intense, and dropped to about 11kHz.
The right channel (RED: speaker with the epoxy coated cone and drive rod has also reached a higher frequency than the left (Green) channel due to the increased stiffness.
At the same time, the values of the right channel in the range below 820Hz also have -3dB less power.
So far, the expectations have been met.
Unclear to me is the Woppel (blue circle) at 300Hz???
Does anyone have an explanation?
Here also the comparison:
Loudspeakers with epoxy resin impregnated coil spiders vs loudspeakers with metal ring
The main resonance frequency has increased from 280Hz to about 820Hz!
The "secondary resonance" has increased from 1.2kHz to 1.8kHz.
The resonances (~9kHz) have been somewhat attenuated.
The resonance at about 10kHz has become more intense, and increased to about 11kHz.
The high frequency range from 12kHz has "calmed down"
I expect, when I receive the new resonant speakers (with harder metal ring), to see a similar improvement in the main resonance with these metal ring versions.
Further search for the solutions for resonance frequencies
To find out if the drive rods cause the resonances in the 12kHz range, I disassembled the speakers again and mounted small dampers on the drive rods.
These dampers are made of silicone and have a weight of about 0.5g per speaker (2 pieces per speaker).
What influence will this intervention have on the frequency image?
What do I expect from this intervention?:
First, a small decrease in the frequency of the main resonance because the coils now have to move more weight.
Second, some reduction in resonances in the higher frequency range (above 6kHz to 15kHz).
I suspect that resonances in the higher frequency range (above 6kHz to 15kHz), can only come from small thin components. I further suspect that my brass bronze rod, 5mm long, has a much higher resonant frequency than the carbon cones, which are softer and larger. These cones are more likely to be below the 10kHz (perhaps below 3kHz, causing the spurious resonance at 1.8kHz?). I have no idea about the carbon rods, which are 25mm long and 3mm in diameter.
I am very curious about the measurement results myself.
See you soon, best regards from Austria
Thomas
You do not have the required permissions to view the files attached to this post.