Looks like you now have before/after plots.
@Tom
Jan's plot show the problem as a low modulation index FM spectra the pitch inaccuracy is just a DC offset all the eccentricity issues can be extracted from the spectra, though an exact analysis would require Bessel functions.
EDIT - The plots actually show histograms but the point is there are only two frequencies at work here the ~3kHz tone on the LP shifted by the pitch error and FM sidebands at .55Hz on that tone. If you could reference a starting point on the reference tone band you could extract the exact center of the record from the phase and amplitude of the sidebands (assuming you could process out the noise).
Last edited:
Hi,
If the eccentricity does differ between each side that scuppers a simple solution,
like expanding the hole to be concentric then and gluing in say an o-ring.
My best suggestion. First open up the hole just enough to fix the eccentricity.
(A standard reamer from both sides with easily do this.)
Then knock up a disk, or a record weight that can be used either way up
that has fits over the spindle and has a pin on its surface, near the hole.
I haven't worked out the details of how to do it accurately, but
all the record would then need is 2 pin sized holes drilled, 1
for each side and the correct hole marked on each side.
How it works should be obvious. My reasoning is the minimum
modification to the original hole, and the minimum modification
of the label, so not rendering the record useless for anyone else.
rgds, sreten.
2 pins, another near the label edge and not in a straight line,
(so 4 label holes) would need no marking and not depend
on the centre hole size, potentially a lot more accurate.
If the eccentricity does differ between each side that scuppers a simple solution,
like expanding the hole to be concentric then and gluing in say an o-ring.
My best suggestion. First open up the hole just enough to fix the eccentricity.
(A standard reamer from both sides with easily do this.)
Then knock up a disk, or a record weight that can be used either way up
that has fits over the spindle and has a pin on its surface, near the hole.
I haven't worked out the details of how to do it accurately, but
all the record would then need is 2 pin sized holes drilled, 1
for each side and the correct hole marked on each side.
How it works should be obvious. My reasoning is the minimum
modification to the original hole, and the minimum modification
of the label, so not rendering the record useless for anyone else.
rgds, sreten.
2 pins, another near the label edge and not in a straight line,
(so 4 label holes) would need no marking and not depend
on the centre hole size, potentially a lot more accurate.
Last edited:
Hi,
Further thought indicates my original idea of two pins works,
whilst the attempt to try to simplify it to one pin doesn't.
You need two pins so hole size doesn't matter.
The hole size is dependant on the most eccentric side,
and therefore by definition is too big for the other side.
rgds, sreten.
Further thought indicates my original idea of two pins works,
whilst the attempt to try to simplify it to one pin doesn't.
You need two pins so hole size doesn't matter.
The hole size is dependant on the most eccentric side,
and therefore by definition is too big for the other side.
rgds, sreten.
@Scott Wurcer,
I understand the diagrams of jan.didden. But I had something else in mind about the unbalance of the frequencies of vinyl records. The diagram below shows the consequences (the highest note of a piano is c''''').
Conclusion: we can clearly hear the unbalance when the eccentricity of the vinyl record is too much. Therefore, the RIAA standard (max 127 mm) is not adequate. A maximum eccentricity of ~0,3 mm have to be the limit.
I understand the diagrams of jan.didden. But I had something else in mind about the unbalance of the frequencies of vinyl records. The diagram below shows the consequences (the highest note of a piano is c''''').
An externally hosted image should be here but it was not working when we last tested it.
Conclusion: we can clearly hear the unbalance when the eccentricity of the vinyl record is too much. Therefore, the RIAA standard (max 127 mm) is not adequate. A maximum eccentricity of ~0,3 mm have to be the limit.
@scott wurcer, @sreten,
The "big trick" is not the (electronic) measuring of the eccentricity or the new alignment of the calibrated record in relation to the centre of the platter. It is the proces of physical centring itself that must be very easy.
Try to imagine you are centring a record without 2 non-concentric rings. You use a laser and you observe the tolerance when you rotate the record around. Now you will push the record at the brim a bit more centric by... hand. How will you apply just enough force to move the record about 0,2 or even 0,1 mm? You can't! So now starts the troublesome job of correcting all your attempts to centre the record with small pushes...
The idea of @sreten - 2 pawls into the puck (it must be 3 pawls) - is a nice idea. Unfortunately, I had this idea about 6 month before and I stopped because there were to much problems to realize the idea (although I bought all the materials).
First, there is no fast and easy way to centre the records (like the 2 non-concentric rings). Second, you have to drill holes into every record so these holes must have exactly the same radius from the centre of the spindle. Therefore, you have to DIY some kind of drilling machine with 3 drills already positioned... I have drawn the constructionm of such a machine and after a while I had to conclude this idea was not practical (too difficult to make, too tyroublesome to handle and... expensive).
Now look to the diagram below. The left diagram is a Nakamichi self centring turntable. The test frequency is 3150 Hz and the deviation is about 4 Hz. Look to the diagram above (my post before) and draw a horizontal line at 3150 Hz. Now a vertical line at 4 Hz and it is clear: the Nakamichi is centring the record with an accuracy of about 0,2 - 0,3 mm (the test record is 7 inch so the track is a bit like the innermost groove).
Using a laser to measure the eccentricity has the advantage of the "unlimited" magnification of the eccentricity (just move the graduated scale further from the laser). The Nakamichi turntable has only a small volume for the calibrating mechanism, thus an eccentricity of about 0,1 mm is technical possible.
The "big trick" is not the (electronic) measuring of the eccentricity or the new alignment of the calibrated record in relation to the centre of the platter. It is the proces of physical centring itself that must be very easy.
Try to imagine you are centring a record without 2 non-concentric rings. You use a laser and you observe the tolerance when you rotate the record around. Now you will push the record at the brim a bit more centric by... hand. How will you apply just enough force to move the record about 0,2 or even 0,1 mm? You can't! So now starts the troublesome job of correcting all your attempts to centre the record with small pushes...
The idea of @sreten - 2 pawls into the puck (it must be 3 pawls) - is a nice idea. Unfortunately, I had this idea about 6 month before and I stopped because there were to much problems to realize the idea (although I bought all the materials).
First, there is no fast and easy way to centre the records (like the 2 non-concentric rings). Second, you have to drill holes into every record so these holes must have exactly the same radius from the centre of the spindle. Therefore, you have to DIY some kind of drilling machine with 3 drills already positioned... I have drawn the constructionm of such a machine and after a while I had to conclude this idea was not practical (too difficult to make, too tyroublesome to handle and... expensive).
Now look to the diagram below. The left diagram is a Nakamichi self centring turntable. The test frequency is 3150 Hz and the deviation is about 4 Hz. Look to the diagram above (my post before) and draw a horizontal line at 3150 Hz. Now a vertical line at 4 Hz and it is clear: the Nakamichi is centring the record with an accuracy of about 0,2 - 0,3 mm (the test record is 7 inch so the track is a bit like the innermost groove).
An externally hosted image should be here but it was not working when we last tested it.
Using a laser to measure the eccentricity has the advantage of the "unlimited" magnification of the eccentricity (just move the graduated scale further from the laser). The Nakamichi turntable has only a small volume for the calibrating mechanism, thus an eccentricity of about 0,1 mm is technical possible.
Last edited:
@Mark Whitney,
Yes, a very nice and clarifying video. I know the process of making CD's. My brother worked for Phonogram (Philips).
Yes, a very nice and clarifying video. I know the process of making CD's. My brother worked for Phonogram (Philips).
We could do it all electronically by extracting the 0.55 Hz modulation and use it to modulate a (direct) drive motor. In this way, the modulated motor speed can exactly compensate the modulation due to eccentricity.
I'm not sure how to get the correct phase data and couple that to the drive modulation, but I'm sure Scott does 😉
Ehhh... did I just blow a promising patent?
Edit:you can do it closed loop, in a loop that regulates the 0.55 Hz modulation towards zero. No other mod needed to either the record or the player.
Actual music content doesn't matter because that will surely not intervene with the 0.55 Hz modulation tone, which can be filtered out just for the control loop.
Jan
I'm not sure how to get the correct phase data and couple that to the drive modulation, but I'm sure Scott does 😉
Ehhh... did I just blow a promising patent?
Edit:you can do it closed loop, in a loop that regulates the 0.55 Hz modulation towards zero. No other mod needed to either the record or the player.
Actual music content doesn't matter because that will surely not intervene with the 0.55 Hz modulation tone, which can be filtered out just for the control loop.
Jan
Last edited:
@jan.didden,
Audiophiles don’t have direct drive turntables. They all have high end turntables with drive belts and very heavy platters. At the other hand: do the owners of Technics turntables care for the distortion by eccentricity? Mostly they use MM cartridges instead of high end MC cartridges. So I suppose the group interested vinyl lovers with a direct drive platter is not so big.
Nevertheless, the idea is really nice and I think there will be a market too, when the result is perfect. Anyway, you have to develop a high end turntable with a direct drive platter to get a lot of people enthusiastic (and it will last a couple of years before the idea/product is well-known).
Filtering a frequency out of “noise” isn’t new. Regulating/correcting the rotational speed of the (light weighted) platter on the fly isn’t new too. B&O did it for the Beograms long, long ago. Nevertheless, give it a try because it is a beautiful idea.
Audiophiles don’t have direct drive turntables. They all have high end turntables with drive belts and very heavy platters. At the other hand: do the owners of Technics turntables care for the distortion by eccentricity? Mostly they use MM cartridges instead of high end MC cartridges. So I suppose the group interested vinyl lovers with a direct drive platter is not so big.
Nevertheless, the idea is really nice and I think there will be a market too, when the result is perfect. Anyway, you have to develop a high end turntable with a direct drive platter to get a lot of people enthusiastic (and it will last a couple of years before the idea/product is well-known).
Filtering a frequency out of “noise” isn’t new. Regulating/correcting the rotational speed of the (light weighted) platter on the fly isn’t new too. B&O did it for the Beograms long, long ago. Nevertheless, give it a try because it is a beautiful idea.
Hi,
Another way of doing it is an eccentric vertical pivot in the spindle,
a clamp / weight with a slot that fits to the top via a slot, and limit
stops in the spindle to limit it to 180 degrees rotation.
(Well at least give you your zero stop.)
Clamp / Weight on the spindle full anti clockwise is no eccentricity.
Rotating the clamp / weight 180 degrees moves it to full eccentricity.
Assuming the records hole fits the spindle and the records
eccentricity is with the range of the vertical offset, then
I can see no reason why via marks on the clamp and
the outside edge of the label will not fix the problem.
The records, other than marks each side, are untouched.
It fixes the problem of "nudging" the record, you do that by
rotating the clamp / weight and the record at the same time,
once you have established the axis of eccentricity and marked it.
You then advance both marks together until you hit concentricity.
You then reverse the clamp / weight to zero and mark its point.
It is then very trivial to align the record properly, in use.
rgds, sreten.
Goes to show your ideas a day later are much better ...
But you've got to think hard the day before ...
Another way of doing it is an eccentric vertical pivot in the spindle,
a clamp / weight with a slot that fits to the top via a slot, and limit
stops in the spindle to limit it to 180 degrees rotation.
(Well at least give you your zero stop.)
Clamp / Weight on the spindle full anti clockwise is no eccentricity.
Rotating the clamp / weight 180 degrees moves it to full eccentricity.
Assuming the records hole fits the spindle and the records
eccentricity is with the range of the vertical offset, then
I can see no reason why via marks on the clamp and
the outside edge of the label will not fix the problem.
The records, other than marks each side, are untouched.
It fixes the problem of "nudging" the record, you do that by
rotating the clamp / weight and the record at the same time,
once you have established the axis of eccentricity and marked it.
You then advance both marks together until you hit concentricity.
You then reverse the clamp / weight to zero and mark its point.
It is then very trivial to align the record properly, in use.
rgds, sreten.
Goes to show your ideas a day later are much better ...
But you've got to think hard the day before ...
Last edited:
I think there is some poetic justice in making all those über-heavy super-expensive platters worthless in one go 😉
But seriously, in hindsight, going for super-stable platter speed seems the wrong way to go in this light. You can speed-modulate a DC motor of course - can you speed-modulate an AC motor easily?
Jan
@sreten,
Well, try it with a piece of cardboard, scissors and a pair of compasses. You will be surprised.
@jan.didden,
You are right, those heavy turntables are a bit overdone. The app of "dr. Feickert" shows it clear: the "fat turntables" have no better rotational speed (even worse). That's because it is hard to control a heavy weight (simple classic physics).
Manipulating the rpm of a AC motor is possible by frequency and voltage level. The latter is easier.
I have searched for the stability of a direct drive platter (Technics SP10 MK2).
The image below shows the result (far better than very heavy platters).
Well, try it with a piece of cardboard, scissors and a pair of compasses. You will be surprised.
@jan.didden,
You are right, those heavy turntables are a bit overdone. The app of "dr. Feickert" shows it clear: the "fat turntables" have no better rotational speed (even worse). That's because it is hard to control a heavy weight (simple classic physics).
Manipulating the rpm of a AC motor is possible by frequency and voltage level. The latter is easier.
I have searched for the stability of a direct drive platter (Technics SP10 MK2).
The image below shows the result (far better than very heavy platters).
An externally hosted image should be here but it was not working when we last tested it.
You can do it with a sound card and PYTHON. I am in Florida or I could demonstrate, I have a recording of an entire RCA STR100 at 24/96 somewhere.
@Tom all this is well and good, but practically speaking I only have a few LP's where it might matter. Since I am not a record collector reaming out the hole a little is no problem for me. Then a laser point across the room, one revolution to find the min/max and then center. Can't take more than a couple of iterations.
@jan.didden,
You can find the link to “dr. Feickert’s webside here. I have bought the test record (17 euro), the app is free (download).
@scott wurcer,
Centring your eccentric records is not important when you have:
· a fixed pivot tonearm (that’s “rubbish”);
· an MM cartridge;
· records with a lot of loud music (much instruments, voices, etc.).
Unfortunately, I have made myself a new type high end linear tonearm. It has improved the quality of sound so much that centring eccentric records “has become a must”. Besides that, I have made a prototype of a new type phono cartridge too and I don’t want a to and fro moving cartridge any more. That’s a nasty relic out of the second part of the century before.
Nevertheless, the size of a vinyl record hole is 7,2 mm and that’s far too small (contact area between record and spindle). It wears too much and it increases the eccentricity of the record. If vinyl recordings are outdated, it has no sense to spend much time to increase the quality of the reproduced sound. But there is a vinyl revival and it don’t stops... Conclusion: improving vinyl isn’t a playground but a serious activity that will be at the good of many people.
You can find the link to “dr. Feickert’s webside here. I have bought the test record (17 euro), the app is free (download).
@scott wurcer,
Centring your eccentric records is not important when you have:
· a fixed pivot tonearm (that’s “rubbish”);
· an MM cartridge;
· records with a lot of loud music (much instruments, voices, etc.).
Unfortunately, I have made myself a new type high end linear tonearm. It has improved the quality of sound so much that centring eccentric records “has become a must”. Besides that, I have made a prototype of a new type phono cartridge too and I don’t want a to and fro moving cartridge any more. That’s a nasty relic out of the second part of the century before.
Nevertheless, the size of a vinyl record hole is 7,2 mm and that’s far too small (contact area between record and spindle). It wears too much and it increases the eccentricity of the record. If vinyl recordings are outdated, it has no sense to spend much time to increase the quality of the reproduced sound. But there is a vinyl revival and it don’t stops... Conclusion: improving vinyl isn’t a playground but a serious activity that will be at the good of many people.
Last edited:
Now my memory from back then is very poor, but I do remember the dragon TT being shown on TV (tomorrows world). What I cannot remember is how the dragon measured eccentricity, but I am pretty sure it was a small arm that measured the outside of the record. Can anyone confirm this?
ah yes, that brings back the memory. I need to have a look at what I could do with my Xerxes. But first I need to find a record that is bad in my collection.
This one shows the glass plate and demonstrates before and after.Nakamichi Dragon CT computing turntable 1985 - YouTube
I kinda like my SME 345, but handy I don't need to worry about this eccentricity.
This is where I can be smug. I don't have that problem. I just need to work out a way of making it a few seconds job to put an off centre record in the right place. Sticky tape may end up being the best solution.
- Status
- Not open for further replies.