Karousel

[Spannko]

Great work ThomasOK! It appears that the difference between optimally torqued and otherwise is critical to getting the best out of the Karousel.

Now that you’ve optimised the torque, it would be really interesting to hear a serviced Cirkus deck compared to the optimised Karousel 🤞

[Defender]

thank you for solving this open question Thomas - its still interesting to me how even with the torques on the karousel the opinion is split ... you and Ron like the lower setting - others with good tune dem experience like the higher setting. I understand you can change the torque without dismantling the LP12 in just doing it from the top while holding the Keel or Kore in place?

Actually I think the „real“ winner is Kamala ;)

[anthony]

It has been pointed out to me via a PM that I am responsible for a dereliction of duty in not explaining the last two clips made at Simon's place. I beg forgiveness but will state that there have been extenuating circumstances such as the endlessly drawn out election over here, several turntables to work on this week and a near brush with COVID-19. Before anybody gets worried we are all OK at Overture Audio. But one of the guys here had contact with a couple who came down with COVID-19 six days later. They believed they had gotten it from a relative two days before his visit so we were all concerned as we had all been working together the week before he found out. So we tried to take the best measures with him self-quarantining and getting a test and the other three of us also getting tested. We also informed customers and went back to appointment only with contactless pickup offered. Fortunately yesterday he and I both received negative results, the other two should receive results today but there is no reason now to think they won't be negative as well. So next week we should be back to more normal business although we may keep on an appointment basis since the infection rates are rising in Michigan again as they are through most of the US.

So with that preamble the clips are of Simon's system which is a Klimax LP12 with Karousel and a Koetsu Rosewood Signature cartridge and an Audio Research Reference 2 phono stage. The rest of the system is a Rega Apollo CD player (quite musical for the money), a Fanfare FT-1A FM tuner, Sagatun 1.4 and Tundra 2.5 stereo units and the 62 year old Quads I used at the Axpona Hi-Fi show in 2019 but with black grills and special hardwood side panels/legs that I stained and finished for him. A CablePro Revelation outlet strip (now sadly unavailable), Lejonklou latest power cables, Linn Silvers and 2.48m black K200 plus a Mimer rack complete the setup.

The comparison is between my original torque on the Karousel of .4Nm -2 notches in Clip 1 vs. my latest (and I hope final) torque of 2.6Nm +5 notches in Clip 2. No other changes were made to the system.

Those with eagle eyes will notice that this is a different torque than the 2.6 +3 I had used in the comparison on my system. After a little reconsideration, based in part on only having compared torques on a Klimax LP12 with Urika and Nordost cables, I decided to do some more testing. I started with the store demo Klimax table with Urika and Linn Silvers and, while I still preferred 2.6 +3 to 4.0 -2, there was something about the lower one that still didn't seem quite optimal. So I tried a handful of other torques in the area and the 2.6 +5 is where the music really all came together. I then did the 2.6Nm +3 vs. +5 on my own LP12 with Simon listening and we both felt the +5 really brought it all together. The above clips were then done when I switched Simon's table over to the 2.6 +5 setting which we both felt really improved his system noticeably. I find it interesting and surprising that 4.0Nm -2 sounds quite good (but not the best) on Keel, Kore and Majik sub chassis, but 3.4 is not good on Keel and around 2.6 is not good on Kore and Majik. So my recommended torque for the Karousel to a Keel is now 2.6Nm + 5 notches. However, for the Kore and Majik the torques in the 2.6Nm range just don't work. For them a torque of 3.4Nm even is the best I have found.

There is another thing that has complicated my life and has made the clips I have done over the last month or so on my system suspect. It turns out my Ekos SE/1 is defective. It has excessive vertical friction which sometimes causes it to track to light. Once I noticed it behaving oddly I put it on my gauge and it showed 1.35 grams, which then increased to 1.85 over the period of about 20 seconds. It is at Linn for repair. So I fitted my Ekos 2 and Adikt and plugged it into a Slipsik 7.1. It still sounds wonderful. So no more clips from my system until I get my system back together with the repaired Ekos SE/1 and my Sagatun Monos.

By the way, while I was writing this the AP called Pennsylvania, and hence the election, for Biden. I have tried to avoid politics on here but I can't pretend I'm not very happy.

I cannot pretend I'm not very happy.......trying to work that out!

[V.A.MKD]

Hi ThomasOK,
Glad that you are OK. Stay safe, the most important, have a good health, you, your family, your colleagues ...
Thank you for clearing all about this question, it's really great work ...

[V.A.MKD]

Great work ThomasOK! It appears that the difference between optimally torqued and otherwise is critical to getting the best out of the Karousel.

Now that you’ve optimised the torque, it would be really interesting to hear a serviced Cirkus deck compared to the optimised Karousel 🤞

+1

[ThomasOK]

thank you for solving this open question Thomas - its still interesting to me how even with the torques on the karousel the opinion is split ... you and Ron like the lower setting - others with good tune dem experience like the higher setting. I understand you can change the torque without dismantling the LP12 in just doing it from the top while holding the Keel or Kore in place?

Actually I think the „real“ winner is Kamala ;)

I will mention that Spannko, Matteo and Freddy also all preferred Clip 2. It seems about evenly split but maybe with more liking 1. Yes, it can be done without even taking the bottom off, the sub chassis really doesn't want to move much so just hold the armboard a bit. The main trick is to not lose any oil which requires setting the inner platter/spindle upside down without any tilt. I use the hard foam piece that comes on top of the parts in the Karousel kit as it has a hole in the middle the spindle can fit in.

On Kamala, I agree. I would say that I am even more excited that the ticket with her won than I am that Biden won, if that makes sense.

[Spannko]

The greatest looser has to be King Nigel Farage of Europe. 😂

[Charlie1]

Hi ThomasOK,
Glad that you are OK. Stay safe, the most important, have a good health, you, your family, your colleagues ...
Thank you for clearing all about this question, it's really great work ...

+1
And congrats on your new President!!!

[sktn77a]

Those with eagle eyes will notice that this is a different torque than the 2.6 +3 I had used in the comparison on my system. After a little reconsideration, based in part on only having compared torques on a Klimax LP12 with Urika and Nordost cables, I decided to do some more testing. I started with the store demo Klimax table with Urika and Linn Silvers and, while I still preferred 2.6 +3 to 4.0 -2, there was something about the lower one that still didn't seem quite optimal. So I tried a handful of other torques in the area and the 2.6 +5 is where the music really all came together. I then did the 2.6Nm +3 vs. +5 on my own LP12 with Simon listening and we both felt the +5 really brought it all together. The above clips were then done when I switched Simon's table over to the 2.6 +5 setting which we both felt really improved his system noticeably. I find it interesting and surprising that 4.0Nm -2 sounds quite good (but not the best) on Keel, Kore and Majik sub chassis, but 3.4 is not good on Keel and around 2.6 is not good on Kore and Majik. So my recommended torque for the Karousel to a Keel is now 2.6Nm + 5 notches. However, for the Kore and Majik the torques in the 2.6Nm range just don't work. For them a torque of 3.4Nm even is the best I have found.

Tom:

Just wondering, as the torques seem to vary quite a bit, have you tried anything higher than 4.0? How about 4.5 or 5.0?

[Freddy]

The comparison is between my original torque on the Karousel of .4Nm -2 notches in Clip 1 vs. my latest (and I hope final) torque of 2.6Nm +5 notches in Clip 2. No other changes were made to the system.
...

Thanks Thomas for doing these tests. Quite intersting that optimal torques are different on the different sub chassis. That is not chase on cirkus, right? /All best Fredrik

[lejonklou]

The comparison is between my original torque on the Karousel of .4Nm -2 notches in Clip 1 vs. my latest (and I hope final) torque of 2.6Nm +5 notches in Clip 2. No other changes were made to the system.
...

Thanks Thomas for doing these tests. Quite intersting that optimal torques are different on the different sub chassis. That is not chase on cirkus, right? /All best Fredrik

With Cirkus the optimal torque is different for each model of subchassis. Keel has the highest torque of around 3.8 Nm, while the Cirkus steel subchassis is around 3.4 Nm and Kore a tiny bit lower than 3.4.

[HBE]

The comparison is between my original torque on the Karousel of .4Nm -2 notches in Clip 1 vs. my latest (and I hope final) torque of 2.6Nm +5 notches in Clip 2. No other changes were made to the system.

What do you mean with -2 and +5 notches. Don't you set the torque driver to the correct torque and that's it?

PS. I guess you mean 4Nm and not .4Nm ;.)

[lejonklou]

The comparison is between my original torque on the Karousel of .4Nm -2 notches in Clip 1 vs. my latest (and I hope final) torque of 2.6Nm +5 notches in Clip 2. No other changes were made to the system.

What do you mean with -2 and +5 notches. Don't you set the torque driver to the correct torque and that's it?

PS. I guess you mean 4Nm and not .4Nm ;.)

With the Sturtevant Richmont CAL-36/4 torque tool, you can fine tune the set torque by turning the adjustor in steps, or even half steps.

You won't be able to see on the analogue scale of the tool exactly what the set torque is, only that it for instance is slightly more than 2.6 Nm. We count the notches to share the precise setting.

[HBE]

The comparison is between my original torque on the Karousel of .4Nm -2 notches in Clip 1 vs. my latest (and I hope final) torque of 2.6Nm +5 notches in Clip 2. No other changes were made to the system.

What do you mean with -2 and +5 notches. Don't you set the torque driver to the correct torque and that's it?

PS. I guess you mean 4Nm and not .4Nm ;.)

With the Sturtevant Richmont CAL-36/4 torque tool, you can fine tune the set torque by turning the adjustor in steps, or even half steps.

You won't be able to see on the analogue scale of the tool exactly what the set torque is, only that it for instance is slightly more than 2.6 Nm. We count the notches to share the precise setting.

So when Thomas says -2 notches he set the torque to 4Nm and go two notches back again.
And when he says +5 notches he set the torque to 2.6Nm and continue with 5 notches.
How many notches is it between 2.6 and 2.7Nm on the Sturtevant tool?

Is this possible with the Wera A5 or A6 torque tool that Linn recommend, i.e. do settings between 2.6 and 2.7Nm for instance?

[ThomasOK]

Fredrik already answered some of the questions the same way I would so I won't repeat that. Yes, it was supposed to be 4.0Nm -2, not .4Nm and I have now corrected that. I have not tested more that a little above 4.0Nm (probably about 4.2) as my driver doesn't go higher than that. Also because Linn's original recommendation was 4.0Nm but was then lowered to 3.5Nm. I don't want to go high enough to damage anything and my torque testing has lead in the opposite direction anyway. The Wera drivers I have looked at do not allow as fine an adjustment as the Richmont-Sturtevant does. There are around 10 notches between each set of marks (2.6Nm and 2.8Nm as it is graduated in .2Nm increments) which means a notch is approximately .02Nm and a half notch, which we have occasionally found beneficial, is closer to .01Nm. I don't think the Wera allow that level of precision, and as I mentioned here somewhere, it takes three Wera drivers to cover the same range as the one SR driver. Also I won a Wera in a contest a while back, didn't use it for a couple of years and then took it out when the Karouse arrived to see if it covered a higher range. When I took it out it fell on the kitchen floor (composite tile) and broke into pieces. That's the end of my experience with Wera drivers. I haven't yet installed a Karousel on a Cirkus sub chassis so I don't have a precise torque for that.

[Defender]

2.6Nm +5notches should be 2.7Nm (to go with the example from HBE)
The only issue I have and Thomas knows that one complete turn is 9 notches and I would have loved the tool to rather have 10 notches for a 0.2Nm step.

I have seen Thomas working with long bits probably 50mm to 70mm and have seen Fredrik using short bits like 25mm length - does the length of the bit change the value as there are some bits that allow some torque to reduce torque spikes and I would like to know if that is counterproductive to use those bits (probably hard to understand what I mean)

[sktn77a]

I see the Sturtevant-Richmont 36/4 torque screwdrivers available on this internet. What isn't obvious is how you use this with, say, the 20mm socket needed to tighten the Karousel bearing nut(?) Is this a special socket or a standard 3/8" socket? if the latter, is there a 1/4" hex to 3/8" square adapter?

[V.A.MKD]

2.6Nm +5notches should be 2.7Nm (to go with the example from HBE)
The only issue I have and Thomas knows that one complete turn is 9 notches and I would have loved the tool to rather have 10 notches for a 0.2Nm step.

I have seen Thomas working with long bits probably 50mm to 70mm and have seen Fredrik using short bits like 25mm length - does the length of the bit change the value as there are some bits that allow some torque to reduce torque spikes and I would like to know if that is counterproductive to use those bits (probably hard to understand what I mean)

+1, good question ...

[Defender]

you use a normal 20mm nut with a 1/4 square to 1/4 hex adapter or a 3/8 square to 1/4 hex adapter or even 1/2 square to 1/4 hex adapter depending what your nut has.

I couldnt find a 20mm nut which you can use directly but there might be in your country.

I assume a normal nut will do and that you dont need a deep nut.

Thomas does the bottom one also need a 20mm nut?

[OscarH]

2.6Nm +5notches should be 2.7Nm (to go with the example from HBE)
The only issue I have and Thomas knows that one complete turn is 9 notches and I would have loved the tool to rather have 10 notches for a 0.2Nm step.

I have seen Thomas working with long bits probably 50mm to 70mm and have seen Fredrik using short bits like 25mm length - does the length of the bit change the value as there are some bits that allow some torque to reduce torque spikes and I would like to know if that is counterproductive to use those bits (probably hard to understand what I mean)

+1, good question ...

There will be elastic deformation in the bit corresponding to the applied torque. Bits of different dimensions may have different stiffness, but this should not impact the accuracy of the threaded assembly as far as I can see.

[ThomasOK]

I use long bits (often called power drive bits here) when they are available as they are easier to work with. The short bits fit so far down into the shaft of the driver that you generally have to use pliers to get them out. I also have a set of very small 4mm bits and an adapter I use for things like the grub screw in the arm pillar that holds in the arm cable plug, the tiny torx screws used to hold the armrest/anti-skating platform to the arm and the upper and lower parts of the Ekos SE arm pillar together as well as the screws used to hold in XLR connectors on the Urika, Radikal and Lingo 4 units. There are also very small phillips tips one of which is needed for the wire terminal on the Radikal sensor board, and small flat blade and hex head tips that are useful various places. The torque used on the longer bits is low enough that I doubt elasticity in the bit itself makes any difference in the setting. For the Karousel you need a 6 point 20mm socket for the top and a 6 point 21mm socket for the bottom. They don't need to be deep sockets, I am just using the standard ones that came with a Wera wrench set I bought a couple of years ago. You just need a hex to 3/8" adapter to use them with the Richmont-Sturtevant driver. I have included links to the Wiha kit and adapter below as well as a 3/8" adapter that would be suitable (I purchased mine from a local hardware store when I was in the middle of installing my own Karousel and realized I needed one). If you are going to work on a Keel you also need a Torx Security T10 bit for the ground connection. There are several other places where you need a T10 and the T10 security works fine for them as well. I got mine from some special tool place but the ones in the link below should work fine.

Wiha 4mm small bit set (very handy):
https://www.amazon.com/Wiha-75992-Preci ... 371&sr=8-2

4mm tip to hex adapter:
https://www.amazon.com/Wiha-75802-Inser ... 0DH4QZZ9CQ

Hex to 3/8" adapter:
https://www.amazon.com/Makita-97053-Imp ... =hi&sr=1-4

Torx Security T10 bits:
https://www.amazon.com/VEGA-Security-Bi ... hi&sr=1-24

[Defender]

thank you Thomas - as always very helpfull - I didn’t thought about 4mm bits so thank you for the advice.
I assume the top nut of the Karousel is the more important one for tune as the Karousel comes with the lower one screwed on from factory.
Are the 1.6Nm, 1.8Nm 2.0Nm etc settings on the Sturtevant always landing at the same notch meaning 9 notches are 0.2Nm or is there a shift? I am asking because than I would use the 1.6 +1notch as a benchmark and always start from that as my Sturtevant has no real clear eye
I was actually thinking of drilling two holes besides the mark to have a better view
————
-o—o-
————

[V.A.MKD]

Thank you Thomas.

As always right and in detail answer.

[alan]

Hello,
I apologize . I just changed what I had written at the beginning, with my new Wera that I received yesterday the user manual is not detailed. I have just noticed that the decimal graduations on the rotary dial of the Wera A5 starts 10 notches before 2.5Nm this distorts my estimate that preceded; the 2.5 Nm line on the scale graduated in height is placed alone on the right. This confirms that the 10 notches progress exactly from coupling 0.1Nm per notch, the decimal values displayed add up to 2.5Nm. 2.5+5 notches provide 3Nm 2.5+10notches is 3.5Nm Wera announces error tolerance +/- 4%

Sorry I use a translator French/English

[lejonklou]

thank you Thomas - as always very helpfull - I didn’t thought about 4mm bits so thank you for the advice.
I assume the top nut of the Karousel is the more important one for tune as the Karousel comes with the lower one screwed on from factory.
Are the 1.6Nm, 1.8Nm 2.0Nm etc settings on the Sturtevant always landing at the same notch meaning 9 notches are 0.2Nm or is there a shift? I am asking because than I would use the 1.6 +1notch as a benchmark and always start from that as my Sturtevant has no real clear eye
I was actually thinking of drilling two holes besides the mark to have a better view
————
-o—o-
————

Don't start at 1.6+1 for all settings, it won't work well. What you likely need to do is polish the inside of your "eye", it gets dirty with use. Loosen the tiny hex screw that holds it and unscrew the eye. Then polish the shaft and the inside of your eye.

Before doing this, leave your tool set to a well known torque that you've set by ear, so that you can return to it when remounting the eye. You mention 1.6+1, which is optional for the tonearm height lock screw on several arms. That is what I use as well.

Once you have a clear eye, you can keep it clean in the future by inserting and rotating a piece of white paper between the shaft and the inside of the eye. That usually picks up the dirt that accumulates.

Please note, also, that it's the odd number of steps (9), that permits the use of half steps on the SR CAL 36/4. I wouldn't want one with 10 steps, because you can't rely solely on counting the steps for precise settings of the tool, it's far too analogue and not digital in its settings. You should always start with a visual setting on a number and its corresponding line in the eye, then add any extra steps.