Sansui SR-636 alignment adventures

[PiotrekN]

I went through the process of perfecting the alignment on my Sansui SR-636 and I found out that there is plenty of information on the web regarding this table and the SR-838 that perhaps might be a bit misleading, so I thought I'll post the few observations I had along the way.
The alignment for SR-636 and SR-838 has been discussed in several threads here at AK, in particular the „overhang shortcut” of 53mm. I will first present my alignment procedure and discuss some observations.

What I consider certain is the following data found in the manuals for both 636 and 838.
The tonearm has:
effective length = 230mm
overhang = 16.1 mm
offset angle 22.5.

(When you check the manuals or brochures for these TTs you’ll find that they all list the data also in inches, in particular overhang is 11/16”, which is 17.46 mm. This seems to be a typo, as crude measurement I made for the spindle-to-pivot distance seems to be indeed 213.9mm rather than 212.54mm.)

Entering the above data into the vinyengine tonearm alignment calculator (https://www.vinylengine.com/tonearm_alignment_calculator_pro.php) yields custom null points:

Inner = 63.517
Outer = 112.517

This geometry should produce much better results in terms of distortion (Arm 1 is the Sansui 636 arm with the above data) than any other standard alignment, at least according to the VE calculator:



For these two points I prepared an arc protractor, with the grids for these null points. The protractor was prepared in LaTeX, a highly accurate typesetting language.
The protractor (A4 paper format) is attached at the bottom of this post.

It was then printed on a transparency and I verified it by hand: by using my son’s compass, drawing an arc of 230 mm and checking that the two arc do indeed agree precisely.
The Sansui SR-838 has the exact same tonearm parameters, according to official Sansui materials, so the same protractor applies.

I aligned the cartridge as perfectly as I could to this arc protractor: the stylus perfectly follows the arc and on the grid at the null points the cartridge and the cantilever are perfectly aligned with the grids tangent to the null point circles. (As one of the aids I used the pencil leads, it was quite helpful). I think I got to the point where I am not able to improve it without a microscope or some similar gadget.





Now, one can find other methods, such as a „shortcut” to set it a’la the Technics gauge to 53mm. This does not seem to be correct and would yield a longer pivot to stylus distance in the case of these tables. The method described above gives the "shortcut" of approx. 51mm, measured with the Denon DL-110 overhang gauge.

The other issue is that the Sansui manual suggests to use the original overhang gauge and align the end of the stylus with a mark on the headshell:




I would assume that this setting should provide the correct stylus-to-pivot distance and overhang. However, on my headshell this also does not work and it would produce overhang larger than 16.1mm.

There are some possible explanations I am considering, as it might be the case that my headshell came from an SR-737, which has different overhang and effective length - if the headshells had these marks customized to the model of the turntable they came with, mine could be off. If not then something is wrong with the Sansui instructions for this, however it is also the case that the cartridge these TTs came with, the SV-43 (rebadged AT-10) had a conical stylus, so perhaps they didn’t care too much. In any case this seems odd.

I would be happy to hear comments or corrections to the above.

 

[PiotrekN]

By the way, as an additional comment, below is a table with all the data calculated by the VE Tonearm Calculator. Again, "Arm 1" is the Sansui published data of the Mass Concentrated Fulcrum tonearm on the SR-636 and SR-838.
The only fixed parameter here is the pivot-2-spindle distance of 213.9mm.
The different alignments correspond to different overhang settings: the smallest for Stevenson 15.962 and the largest for Lofgren B with 18.369, over 2mm more than the tonearm spec.
The effective length (stylus-2-pivot) changes from 229.862mm to 232.269mm, respectively.
The offset angle also changes for these alignments, suggesting that an appropriate, very slight twist of the cartridge in the headshell by approximately 1.2 deg is needed to achieve the desired result for the Lofgrens.

It seems to me that Lofgren B or Lofgren A (Baerwald) are close to the mark on the headshell and the Sansui setting from the manual for the SR-636 and SR-838.
On the other hand, the alignment forced by the MCF tonearm data is closer to Stevenson. This often the case for Japanese tonearms from that period.

 

[Blue Shadow]

I like the Baerwald, Lofgren A in your chart and diagram because it makes the maximum error equal at the lead-out, center and lead-in areas of the modulated groove. Center being the area between the two null points not the exact center of the record or exact center of the distance between the null points. Some may like the Stevenson for it's low tracking error on the inner grooves but I don't experience IGD.

Nice work putting this out there. I don't think the Japanese table manufacturers were as concerned with alignment as we are today. Some of the alignments were nicely done and others just bolted the cartridge to the headshell and off we go, playing the record. A 1980? article in one of the audio rags did testing of a bunch of tonearms and the tracking errors were all over the place. There were three arms that has low tracking error for all three test points, seemingly following the Baerwald alignment.

I had the Feikert here for a time and Mike brought over his tonearm/armboard, used on the same table as mine, a TD-125 and we moved the needle less than 1 millimeter and it made a noticeable improvement in the sound. So all this fiddling with getting the arm set up to a proper alignment is good for the playback system. Of course, you need to decide which alignment to use and then try to accomplish that with your set up as was done here.

 

[Vintagear]

I previously had the carts on my 838 aligned with the MoFi Geodisc which I believe was very close to a Baerwald protractor. I'd have to double check but I think that when I re-aligned the carts with the Sansui Gauge and marker point on the headshell it lined up perfectly with a Stevenson protractor. With the latter, the stylus was not as far forward and I felt that it sounded better, more open.

 

[PiotrekN]

I like the Baerwald, Lofgren A in your chart and diagram because it makes the maximum error equal at the lead-out, center and lead-in areas of the modulated groove. Center being the area between the two null points not the exact center of the record or exact center of the distance between the null points. Some may like the Stevenson for it's low tracking error on the inner grooves but I don't experience IGD.

Nice work putting this out there. I don't think the Japanese table manufacturers were as concerned with alignment as we are today. Some of the alignments were nicely done and others just bolted the cartridge to the headshell and off we go, playing the record. A 1980? article in one of the audio rags did testing of a bunch of tonearms and the tracking errors were all over the place. There were three arms that has low tracking error for all three test points, seemingly following the Baerwald alignment.

I had the Feikert here for a time and Mike brought over his tonearm/armboard, used on the same table as mine, a TD-125 and we moved the needle less than 1 millimeter and it made a noticeable improvement in the sound. So all this fiddling with getting the arm set up to a proper alignment is good for the playback system. Of course, you need to decide which alignment to use and then try to accomplish that with your set up as was done here.

Indeed, what the "best" alignment depends on several factors, most importantly what kind of records does one have/play. In my case I have plenty of jazz albums that are cut quite hot, run close to the label and more often than not the ending of a piece involves all the instruments going off loudly together. Thus my choice of alignment and choice of cartridge - the Denon picture is outdated, I am using an AT VM95ML with a MicroLine stylus (which is much better than the Denon, BTW). In my case also a tiny adjustment made a huge difference. It pays off to do it properly!

The one thing that I find odd is that the published tonearm geometry and the overhang adjustment method in the manual give different alignments. Seems sloppy more than intentional, but perhaps, as you say, they didn't care as much back in the day.

 

[PiotrekN]

I previously had the carts on my 838 aligned with the MoFi Geodisc which I believe was very close to a Baerwald protractor. I'd have to double check but I think that when I re-aligned the carts with the Sansui Gauge and marker point on the headshell it lined up perfectly with a Stevenson protractor. With the latter, the stylus was not as far forward and I felt that it sounded better, more open.

Is the mark on your SR-838 headshell at the same spot as the one I posted above? I would be curious to know.

 

[Vintagear]

Is the mark on your SR-838 headshell at the same spot as the one I posted above? I would be curious to know.

Yes, I have four original headshells from that series and it seems that there are at least two variations but the only difference I can see is the collar around the azimuth adjustment screw as seen here.



Otherwise the headshells are identical in length and the marker point in the same spot.



The angles in these pics are not perfect but you can get the idea. BTW, I just checked one of the carts and it's definitely a Stevenson alignment. I don't have a way to properly measure the overhang (center of spindle to stylus tip) so I'm not sure what it works out to on these.







Tip: If you are using the Sansui Gauge make sure that when you are looking at the side of the headshell perfectly so that the front of the cartridge body is perpendicular to you line of sight (no yaw). In other words, that you are directly looking at the side of the cart. Also, the top single marker line on the Gauge is an indicator for stock Sansui carts and may work with some carts and not others so you can disregard it for non-Sansui (AT) carts and just follow the line from the marker point on the headshell down the gauge to get the recommended stylus tip position.

 

[Vintagear]

Now, one can find other methods, such as a „shortcut” to set it a’la the Technics gauge to 53mm. This does not seem to be correct and would yield a longer pivot to stylus distance in the case of these tables. The method described above gives the "shortcut" of approx. 51mm, measured with the Denon DL-110 overhang gauge.

Interestingly, if I put any of the headshells/carts pictured above on a Techncs 52mm alignment tool the stylus is forward (longer) than the 52mm point.





,

 

[Vintagear]

Update:

@PiotrekN since you put a lot of work and thought into this great post I printed the protractor you provided and checked it against one of the carts I had already aligned using the Sansui Gauge (same cart as seen in previous pics) and the results were spot on. Both points and the arc were as expected. This makes sense to me because as I said the Sansui Gauge gave me a Stevenson alignment and your protractor is as you said closer to a Stevenson.

I suspect that you did something wrong when you used your gauge since you said you thought it was closer to a Lofgren/Baerwald alignment. If used properly the Sansui gauge will match (or be very close) to the VE Stevenson protractor and the 636 protractor you created. Nice work on the protractor BTW.


Here are some pics of a cart I had aligned with the Sansui Gauge riding on your protractor. Looks like the Gauge works pretty well.

 

[PiotrekN]

Let me start by saying WOW!!! You have four of these headshells! They are quite rare and certainly not easy to find. I am happy that I have one -- but four! Nice!!!

As for the alignment, here are some of my photos. As you can see the mark on the headshell and the tip of the stylus are not aligned. Estimating the distance from the ring on the connector I get 52.5mm to the mark on the headshell, while the distance to the tip of the stylus is approx. 51mm measured with the Denon gauge. So the difference is not that small. The length of my headshell including the connector ring is 58mm.
I don't have the original Sansui gauge, my understanding however is that an imaginary line through the stylus tip and the mark on the headshell should be perpendicular to the bottom surface of the headshell. This is also the impression I get from the photos you posted - but perhaps the gauge works slightly differently, e.g. that angle should be not the right angle?



As for the protractor, your pictures do look approximately correct, however I can't say more. I do have the impression that on a couple of them the stylus might not be exactly on the arc and that the headshell might be tilted in relation to the grid at the null points, but it very well could be perspective and a better photo would be necessary.

On my table I can only assess this when looking from the side and using a flashlight.
Here is what I get, the stylus is visible to be exactly on the arc at the edge of the record and then at the furthest possible point where the arm is still held by the lifter. Of course, the arc cannot be moved between these measurements; the line from the spindle to the pivot on the protractor should point exactly to the pivot point. I also use a damaged LP underneath to make sure all this takes place at the same height as on a record.





I forgot to mention in my original post that it is easy to introduce a non-trivial error when cutting out the spindle hole. For this reason the protractor I posted has two lines tangent to the spindle hole, just cut along them with an exacto knife in the pattern marked in red below. This will make a "triangular flap" and the two tangent point will hold the protractor in the correct position via the two points tangent to the red lines and the tension from the "flap". As far as I can tell this procedure is much more precise than cutting out an actual circular spindle hole.

 

[Vintagear]

Let me start by saying WOW!!! You have four of these headshells! They are quite rare and certainly not easy to find. I am happy that I have one -- but four! Nice!!!

As for the alignment, here are some of my photos. As you can see the mark on the headshell and the tip of the stylus are not aligned. Estimating the distance from the ring on the connector I get 52.5mm to the mark on the headshell, while the distance to the tip of the stylus is approx. 51mm measured with the Denon gauge. So the difference is not that small. The length of my headshell including the connector ring is 58mm.
I don't have the original Sansui gauge, my understanding however is that an imaginary line through the stylus tip and the mark on the headshell should be perpendicular to the bottom surface of the headshell. This is also the impression I get from the photos you posted - but perhaps the gauge works slightly differently, e.g. that angle should be not the right angle?

View attachment 1684347

As for the protractor, your pictures do look approximately correct, however I can't say more. I do have the impression that on a couple of them the stylus might not be exactly on the arc and that the headshell might be tilted in relation to the grid at the null points, but it very well could be perspective and a better photo would be necessary.

On my table I can only assess this when looking from the side and using a flashlight.
Here is what I get, the stylus is visible to be exactly on the arc at the edge of the record and then at the furthest possible point where the arm is still held by the lifter. Of course, the arc cannot be moved between these measurements; the line from the spindle to the pivot on the protractor should point exactly to the pivot point. I also use a damaged LP underneath to make sure all this takes place at the same height as on a record.

View attachment 1684348

View attachment 1684349

I forgot to mention in my original post that it is easy to introduce a non-trivial error when cutting out the spindle hole. For this reason the protractor I posted has two lines tangent to the spindle hole, just cut along them with an exacto knife in the pattern marked in red below. This will make a "triangular flap" and the two tangent point will hold the protractor in the correct position via the two points tangent to the red lines and the tension from the "flap". As far as I can tell this procedure is much more precise than cutting out an actual circular spindle hole.

View attachment 1684354

Just to be 100% sure I printed another protractor and cut the opening as you indicated with an exacto knife. Here are some better side angle pictures of the stylus tip on the arc. This time I used the Sansui (Audio Technica) cartridge because it has more side-view clearance so that I could get a better shot of the stylus tip. As you can see all points on the arc look very, very good.

The only other reason I can think of as to why the Sansui recommended alignment point on the headshell puts you too far forward is that maybe your tonearm base shifted/moved at some point in it's life.



Mid-point between lead groove and 1st null point


1st null point


Mid-point between 1st & 2nd null point


2nd null point


Mid-point between 2nd null point and spindle


As close as cart body would allow me to get to spindle


What setting looks like with Sansui Gauge (two views)

 

[PiotrekN]

What setting looks like with Sansui Gauge (two views)

Very interesting. I can't imagine the tonearm being moved in any way without leaving some marks on the plinth.
In the photos with the gauge, isn't it the case that the stylus tip is in fact shifted back in relation to the mark on the headshell?
It looks from the photos like the the tip of the cantilever is aligned with the mark, but the stylus tip is further back.

 

[Vintagear]

Very interesting. I can't imagine the tonearm being moved in any way without leaving some marks on the plinth.
In the photos with the gauge, isn't it the case that the stylus tip is in fact shifted back in relation to the mark on the headshell?
It looks from the photos like the the tip of the cantilever is aligned with the mark, but the stylus tip is further back.

The mark on the headshell is just behind ( .25mm?) the #9 line on the gauge and to me it looks like the stylus tip lines up in the same spot (<9). In both shots the cantilever tip appears (at least to me) to be beyond the #9 line (>9). Also, keep in mind that the distance between the lines on the gauge is very, very small, probably 1mm (looks huge in the picture). In any event, in real life it appears to me to line up perfectly with the mark on the headshell.

This will give you an idea of the scale of the gauge vs 1/16".

 

[PiotrekN]

The mark on the headshell is just behind the #9 line and to me it looks like the stylus tip lines up in the same spot (<9). In both shots the cantilever tip appears (at least to me) to be beyond the #9 line (>9). In any event, in real life it appears to me to line up perfectly with the mark on the headshell.

I took a close look at the picture with the four headshells: it seems to me that in the two on the left the ring on the connector is noticeably thinner than in the two on the right. In particular, the Sansui SV-43 cart looks like being mounted on the one with a thinner connector. Is this correct?
My headshell seems to have the connector with the thicker ring.
If this observation is correct then this would explain these anomalies, as the thickness of this ring will certainly influence proper alignment: for the headshell with a thicker ring the cart needs to be further back.

 

[Vintagear]

I took a close look at the picture with the four headshells: it seems to me that in the two on the left the ring on the connector is noticeably thinner than in the two on the right. In particular, the Sansui SV-43 cart looks like being mounted on the one with a thinner connector. Is this correct?
My headshell seems to have the connector with the thicker ring.
If this observation is correct then this would explain these anomalies, as the thickness of this ring will certainly influence proper alignment: for the headshell with a thicker ring the cart needs to be further back.

Good eye, but only the one on the far left without a cart has a thinner collar base. I had already thought of that possibility but the headshell the Sansui cart is mounted on is the same as yours with the same azimuth screw (open side) and the thicker collar ring.

Either way even if I had used the headshell with the thinner collar the difference in thickness is probably less than 1mm.

 

[Vintagear]

Another thought, @PiotrekN have you checked your VTA? Is your tonearm level or are you riding tail down?

 

[PiotrekN]

Another thought, @PiotrekN have you checked your VTA? Is your tonearm level or are you riding tail down?

VTA could not be responsible for this - with a stylus to pivot distance of 230mm a change of 1mm in height at the base of the tonearm corresponds to approx. 1/100mm change in the effective length. However, yes, my tonearm is level, adjusted it not long ago for the new cartridge.
I think this all this just proves what I wrote in the original post, that an arc protractor is the way to go - the various shortcuts and the Sansui overhang gauge might not be the best option for every owner of one these two tables. I'd very much like to know the reason for it though.

 

[Vintagear]

I think this all this just proves what I wrote in the original post, that an arc protractor is the way to go - the various shortcuts and the Sansui overhang gauge might not be the best option for every owner of one these two tables.

Agreed.

I'd very much like to know the reason for it though.

Me too . I'm still wondering about the tonearm base position. I can't see any other reason why the marker on the stock headshell would not align properly on your table.

Just out of curiosity, if you have a standard wooden ruler maybe you can wedge it into the corner of the metal base and left side of fulcrum with arm on rest (as seen below) to see if you get the same approximate distance from that point to the edge of platter and center of the spindle. I did it this way because the ruler will remain level with the platter so it will be a straight line (even though it is not the center of the pivot).

Since the start edge of the ruler has some dead space ( ~1/16" as seen in the first pic) I placed a cm tape above the ruler and I get 19.5cm (195mm) from that outer pivot edge to the center of the spindle.

 

[Vintagear]

@PiotrekN, I have one last observation. If you look at the headshell marker in question it lines up with the bottom edge of the word Sansui printed on the top of the headshell. If you measure a straight and parallel line from the center of the fulcrum gimbal to the bottom-center of the word Sansui printed on the headshell you get ~230mm. It seems to me that this indicates that the marker on the headshell is correct.

Let us know if you take any futher measurements to determine why you have to pull the cart back on your headshell to get the proper alignment and let me know if there is anything you would like me to measure on my end.

 

[PiotrekN]

@PiotrekN, I have one last observation. If you look at the headshell marker in question it lines up with the bottom edge of the word Sansui printed on the top of the headshell. If you measure a straight and parallel line from the center of the fulcrum gimbal to the bottom-center of the word Sansui printed on the headshell you get ~230mm. It seems to me that this indicates that the marker on the headshell is correct.

Let us know if you take any futher measurements to determine why you have to pull the cart back on your headshell to get the proper alignment and let me know if there is anything you would like me to measure on my end.


https://i.imgur.com/0yHP6Sal.jpg[/img]']

Below are my photos - all this can be a bit inaccurate, but it does seem that my tonearm with the headshell is slightly longer than yours.
That is interesting and I wonder what is the reason.