|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
122.59.208.98
In Reply to: RE: Technics SL-1200 Alignment posted by John Elison on February 16, 2017 at 07:08:38
Hi John
"I don't understand what you mean by focusing on the linear offset. The determining factor as to whether the Technics alignment is Stevenson's would be the modulated groove envelope."
Yes, I suspected you would notice that my statement didn't make sense - I actually wrote pretty much the opposite of what I was trying to say! Anyway I had written what you just wrote above previously when talking about the Jelco alignment so you know it was a genuine stuff up. I'm in intense pain with a perforated appendix and finding it decidedly hard to concentrate!
I was half way between two different ideas and was hoping to use Linear Offset as the pivotal idea ...
What I was TRYING to say which you already repeated in your response...is that unless you fiddle the modulated envelope to get the offset and overhang, you end up with a different linear offset to the Stevenson Design optimisations in Table 2 (which of course is obvious based on the null point separation).
I was going to reply to your previous message, but saw your subsequent exchange with Garven - I didn't realise you hadn't read his paper in which case we might have avoided some of our more vociferous exchanges! By now you will have seen for yourself what I have been saying in previous exchanges that Stevenson explicitly states that he is selecting Xinner as a "typical" finishing radius which is greater than Xmin. Table 2 clearly identifies his chosen modulated envelope and shows he was applying Lofgren A and adapting his chosen envelope to balance with record speed (45rpm for 7") and achieving the widest practical envelope without excessively increasing tracking distortion.
This is my Quadtractor. I made it (using a laser cutter) and repurposed the trammel and mounting block from my original Linear Offset jig which was so poorly made that it was unusable for my purposes. The reference pin if from the Pro-Ject Align It. The intention was to enable me to compare any one of four optimisations that I calculated using Lofgren A on my chosen radii: (From L to R) DIN LP radii (LO=91.14), Technics Universal(LO=86.16), IEC LP radii (LO=93.52) and finally my chosen own for 7" and 10" records only (LO=80.26 for an envelope 54 to 120.9mm). Despite its crude appearance, the laser etchings for the grid alignments are within 0.025mm. Now that I have refined my technique, using a USB microscope I can now get within ±0.05mm error on the overhang and better than ±0.3deg error on the offset. The dominant error is still the offset angle.
The short version is that with an elliptical tip, the IEC and DIN optimisations have better focus and a smoother presentation than the Technics on the outermost grooves and the inter-null maximum position. However, on the inner grooves for a 7", the Technics is the clear winner. Of course the ultimate is the 7" and 10" Lofgren A optimisation which has the lowest RMS distortion for those records.
A MicroLine closes the gap somewhat and makes the Technics alignment a very viable "general purpose" alignment although for critical listening the smear is noticeable on the outermost radii until you get <140mm.
The key bit of information that my jig gave me was that the Technics arm has very wide tolerances for effective length (up to 0.5mm error) and offset and that reliance on the L-Jig is likely to give rise to significant errors. This is why I questioned the validity of Fremer's comparison.
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
Follow Ups:
Hi Anthony,
Sorry to hear about your appendix. I'm in a similar boat. I had my right knee replaced last week and I have been in serious pain, too. It has been a week and the pain hasn't let up yet.
I guess I had never actually read Stevenson's paper, but in all the other reading I've done regarding Stevenson's alignment, it's accepted definition is as follows:
"Stevenson's alignment places the inner null-point coincident with the innermost modulated groove and distortion is minimized by placing the outer null-point such that the weighted tracking error curve is minimized. In other words, the two maximum points on the weighted tracking error curve are minimized and equalized. Therefore, if you use the IEC definition for inner and outer modulated groove radii, Stevenson's alignment will have null-points of 60.325-mm and 117.42-mm."
This is basically what everyone believes to be the definition of Stevenson's alignment. Therefore, it doesn't make a lot of sense to talk about Stevenson's alignment differently, especially when your alternate definition hinges on Lofgren A.
My leg is killing me at the moment so I will have to quit. However, I will scrutinize your post more thoroughly and respond more appropriately later.
Thanks,
John Elison
Hi John
Ouch! Knee ops can take a while to settle down pain wise. I presume your doctor/specialist has prescribed you with appropriate relief - the problem is dealing with breakthrough pain.
Just as well everyone knows we are in separate countries, they may have thought the discussion got serious and you fought off an attempted submission attempt by knee-bar!
Seriously though, I wish you well and hope things settle soon.
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
Hi Anthony,
I guess I didn't realize we were in separate countries. I've never been to New Zealand, but my daughter visited for a summer and picked fruit to make money to live. She was having fun traveling around the world.
Anyway, getting back to Stevenson's alignment, what is your simple definition of Stevenson? I gave you mine, which I think is universally acknowledged by most audio enthusiasts.
Another question that might be easier to answer is: What tonearm or tonearms do you know of that incorporate Stevenson's geometry? If you can point me to some tonearms, that would be very helpful.
Thank you,
John Elison
Hi John
I trust that your daughter had a good Kiwi Experience? It is certainly a wonderful country and our geographic isolation seems to have recently become popular amongst people seeking to escape from the political upheavel in their respective continents!
Frank already beat me to it! The only ones I know off the top of my head without doing a search on VE is the SME 3009 Series II, Series III/S with the original wand (which use Design 1B) and the 309/310 and 312 which use Design 1A. For some reason I thought Rega did, but then when I checked the data in VE the geometry is completely off making the use of a standard protractor mandatory. Taking into account calculation rounding errors etc, I don't consider 0.1mm deviation from the "accepted" nulls to make them different especially when it also depends on the frame of reference defined by the radii from which the linear offset is derived after applying Lofgren A.
Now back to our original discussion. I thought I already did provide a description based on his paper! You wrote it off as simply restating Lofgren A derived solutions from the given modulated envelope.
His paper makes a distinction between Xmin and Xinner and it is critically important to note that Xmin != Xinner. Furthermore his Table 2 clearly identifies his Xmin/Xmax and Xinner and as you already acknowledged, the nulls are derived by applying Lofgren A to the data for Xmin and Xmax. As far as the "Universally Accepted" so-called "definition" is concerned, the difference could be considered subtle, but to me it is actually very different and I believe the result of a fundamental misunderstanding of the paper. The key misunderstanding is that the envelope chosen for calculation of the nulls using Lofgren A is NOT for a single record size which is what the Populist view is. The Popular Understanding would have one believe that the only records people play are 12" LPs. In applying his method, the intention is that you choose a null at the "typical" minimum and NOT, I repeat NOT(!!!!!) the absolute minimum Xmin. Furthermore, the Design 1B nulls presented which are supposedly universally accepted as being the so-called "Stevenson" alignment are clearly labelled in Table 2 as applying to an optimisation intended for 3 record sizes.
By your own admission, most people have not actually read his paper and yet purport to understand the material contained within simply because they believe the original source for their information! From my perspective, the "accepted definition" is the equivalent of an Alternative Fact and the fact that it is "universal" does not make it correct or indeed a Fact. I don't deal in Alternative Facts. I simply stick to the Facts which I obtain directly from the Source!
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
Thanks! My daughter enjoyed her Kiwi experience very much. She was there for a year. She picked apples and worked at a ski resort. She even went sky diving in New Zealand. She also visited Australia for about six months and spent a couple of months living on a beach in Fiji. This was years ago just after she graduated from college.
Anyway getting back to Stevenson! I don't believe the SME 309 and 310 are Stevenson. The story I heard was that some people complained about inner groove distortion with the SME IV and V so SME decided to change the modulated groove envelope to move the inner null-point slightly closer to the spindle. Instead of using 146.05-mm for the outermost groove they rounded it off to 146-mm and in order to move the inner null-point inward slightly, they reduced the innermost modulated groove from 60.325-mm to 58-mm. Then they applied Lofgren A, which resulted in alignment null-points of 63.6-mm and 119.5-mm as shown in the following graph. As you can see, there are no rounding errors whatsoever for any of the SME specified tonearm parameters. However, the null-points of 63.6-mm and 119.5-mm are not Stevenson null-points as depicted in your table. They might be close, but SME doesn't make mistakes. If the SME had chosen Stevenson null-points, they would be exact and not just close. Both the 309 and 310 use Lofgren "A" alignments.
.
The same applies to the SME 310:
.
With respect to the SME Series II and Series III, I also have problems understanding how they could be Stevenson. The owner's manual for the SME III says its effective length is 229-mm and its mounting distance is 215.4-mm, but no offset angle is given. No matter! There is no offset angle that will produce Stevenson null-points from those specifications. The same situation applies to the SME Series II.
Perhaps I'm missing the boat on the SME II and III, but I can't find any numbers that point to Stevenson. If you have the correct parameters for effective length, pivot-to-spindle, and offset angle for these SME tonearms that will yield any of the Stevenson null-points from your table, please help me. I'm not trying to obstinate; I just can't make the SME numbers work.
One tonearm that I believe might be Stevenson is the Dynavector 507. Its graph is below. Tell me what you think.
Thanks,
John Elison
Hi John
I believe you are absolutely right about the 3009/3012 Mk1 geometry not matching the nulls quoted on VE. I should have fact-checked the VE data first, but I just had a look at one of the manuals and although the resolution is very poor, they explicitly quote a null at what looks like 2.75" (69.85mm). So as far as Design 1B goes, it appears not to match the linear offset of ~88.65mm.
Yes, the Dynavector 507 certainly looks to have been at least inspired by the paper if not actually implementing the specific design. The linear offset you calculated is 88.33mm which is compellingly close to the 88.65mm for Design 1B. In practical terms, anyone aiming for one solution will likely achieve close to the other given the errors involved in alignment. This is why I originally said that I wouldn't split hairs on rounding related differences and why I originally said we should focus on the linear offset.
My view is that it is only sensible to say that one arm uses a Stevenson inspired alignment compared to another if the same modulation envelope is used when comparing the design parameters. Then the linear offset will immediately tell you if that is the case or not.
Using this logic I would concede that the 309/310 are not an exact match to Design 1A. However, the linear offset of 91.54mm (SME) compared to 91.21mm (Design 1A) is very similar suggesting that similar considerations had been made in the optimisation of the design by SME. In fact, no arm would be using the Stevenson optimisations since his Xmax chosen Xmax and Xmin were all derived by empirical means rather than by referencing the limits specified in the relevant standards.
Given the evidence presented in Table 2, would you accept that Stevenson is using Lofgren A (but simply strategically choosing his modulation envelope)?
If so, then I think the only point on which we need to debate is whether you accept that Xmin != Xinner and Xinner > Xmin?
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
Yeah, that arm was before my time. However, I don't think any SME tonearms incorporate Stevenson's alignment and certainly not the 309 and 310. These two are definitely Lofgren A with a modulated groove envelope in-between 58-mm and 146mm. This matches exactly! It's not close; it's exact! Furthermore, the 300 Series are modern tonearms and SME uses Lofgren A exclusively on their modern tonearms from the SME III forward.
The more I read about Stevenson, the more I don't think it applies to modern tonearm. Most of us use 12" LPs almost exclusively and Stevenson doesn't even produce an alignment designed excursively for 12" LPs. He has an alignment dedicated exclusively to 7" records whereas all his other alignments are generally applicable to all different size records combined. Apparently, Stevenson came along at a time when 7" 45s were the main concern.
Anyway, I guess we will just have to be in disagreement about the desirability and validity of Stevenson's alignment methods. I don't think Stevenson's alignment (whatever that means) is applicable today. Furthermore, the only tonearms that I know of that use Stevenson's alignment are the Dynavector tonearms. I'm sure there must be others, but I'm not aware of them at the moment and I don't think SME has ever used Stevenson's alignment. However, I could be wrong!
Best regards,
John Elison
"The more I read about Stevenson, the more I don't think it applies to modern tonearm. Most of us use 12" LPs almost exclusively and Stevenson doesn't even produce an alignment designed excursively for 12" LPs. "
To say that Stevenson is not relevant to the modern tonearm is to ignore that fact that he is using Lofgren A - it is that you just don't agree with his choice of modulated envelope and recommended universal solutiom! Secondly, he is providing a design guide with recommended offset angles and overhang in relation to the effective length (refer to fig 6 on P.315 of his paper) for anyone not wanting to reinvent the wheel if they intend to design an arm with minimum peak weighted distortion.
He does in fact recommend an alignment suited to 12" LPs - Design 1A. It is essentially the same as the 309 but for the slightly reduced maximum radius of modulation from 146 to 145.3mm and VERY marginal reduction in Xmin from 58 to 57.9mm!!!! The fact that he uses Lofgren A SURELY convinces you that Stevenson is NOT defined by one single set of nulls. I urge you to read the footnotes in the context of the extract shown in this snip.
I have mentioned this before to you, but I disagree with the suggestion that 12" LPs are the only discs that people play. 7" singles and 10" singles as well as maxi 12"s (both 45 and 33rpm) are still being produced and I still continue to purchase them so a Universal solution such as Design 1B is still very relevant for someone wanting to avoid making changes to the aligment everytime a different record is played. I am by no means unique in this respect and there are plenty of inmates who have amassed vast collections of all different record sizes.
Now if you had reworded your statement to be that "no manufacturer has designed an arm that is currently in production that uses the design parameters for Stevenson Design 1B" then I would have no issue at all and would happily replied with a suitable "+1 nt" in the subject line! :)
Now that you have read his paper, there is still one point I raised previously that you haven't addressed yet and that is:
Do you disagree that Stevenson makes the distinction between Xinner and Xmin such that Xmin < Xinner and Xmin != Xinner except in the case of Design 1A where he has determined a modulated envelope where the "typical" minimum radius Xinner coincides with where he requires Xinner to be due to the Lofgren A solution for the envelope with Xmin < Xinner. That will sound a bit circular, but all I'm saying is he is applying Lofgren A to an LP envelope. In this case, he very clearly is not placing Xinner at Xmin as everyone erroneously quotes as a so-called Stevenson alignment.
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
> Do you disagree that Stevenson makes the distinction between Xinner and Xmin such that Xmin < Xinner and Xmin != Xinner
> except in the case of Design 1A where he has determined a modulated envelope where the "typical" minimum radius Xinner
> coincides with where he requires Xinner to be due to the Lofgren A solution for the envelope with Xmin < Xinner.
This is kind of confusing. Let me see if I can figure this out.
X 0 = X inner = Inner Null-Point
X 3 = X min = Inner groove for maximum distortion
Normally, we would call X 3 the innermost modulated groove radius, but Sevenson refers to it as a point of maximum distortion, which leads me to believe that he does not consider this point part of the modulated groove envelope. It's simply a point that yields the inner null-point when applying Lofgren's "A" method for minimizing tracking error distortion. Furthermore, this point is quite a bit smaller than would be expected for the innermost modulated groove radius on modern LPs. I think this is why nearly everyone except you believes the innermost modulated groove is coincident with Stevenson's inner null-point.
You wrote: Xmin != Xinner except in the case of Design 1A... I don't know what you mean by Xmin != Xinner
All I can say is that according to Stevenson, X min never equals X inner because X min is a point of maximum distortion and X inner is a null-point with zero distortion. However, nearly all of us believe that the inner null-point is coincident with the innermost modulated groove for Stevenson's alignment.
Actually, this is much too confusing. I wonder if anyone else understands what we're talking about. I'm not sure I fully understand. It's just so much simpler for me to think of Stevenson's alignment being the same as Lofgren's "A" alignment with the caveat that the inner null-point be coincident with the inner most modulated groove radius. I'm sorry, but that is how I view Stevenson's alignment.
The Dynavector tonearms use Stevenson's geometry and possibly some older SME tonearms also use Stevenson's geometry. I'm not fully convinced about SME, though.
I apologize for not addressing all your points and questions, but I'm just not thinking clearly at the moment and my leg is throbbing.
Sorry,
John Elison
"I don't know what you mean by Xmin != Xinner"
Sorry! My bad - I get in the habit of using software operators as shorthand for logical operators the way I would in C programming.
"!=" means "Not Equal to" in C
I do hope your knee pain starts to settle down - I know the frustration of continual pain all too well.
Take it easy.
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
Thanks! My knee is still throbbing but I think the pain is subsiding a little.
I found a DIN specification in my picture gallery with some interesting numbers but it's in German and I don't know what it says. It specifies innermost and outermost modulated groove diameters. When I divide these number by 2 to reflect groove radius, I get the following:
Outermost modulated groove radius = 146.3-mm
Inner modulated groove radius = 63.5-mm
Another inner modulated groove radius = 57.5-mm
These are from the last three blocks on the first page of the DIN specification. Perhaps someone who understands German can provide the definition of the last two numbers. I thought the 63.5-mm radius was interesting because Stevenson's 1A Design contains that number for X 0 .
.
.
The DIN standard specifies the minimum radius as 57.5mm. It isn't a hint that Stevenson is considering an inner null at the minimum radius. However, we would need a cutting engineer to tell us what Best Practice is with respect to a "Typical" finishing radius relative to the playing time and "loudness" required. However, the 63.5mm is most definitely not a minimum value.
The majority of my pop LPs from the 80s from Germany, Britain, New Zealand are all cut in further than 60mm with 58mm being very common. Interestingly I have some that appear to have an extended lead in spacing so that the track starts at a radius of ~142 to 143mm, but is cut in to 58mm (both German and British examples). Odd that they would choose to push inwards towards the inner groove limit than to keep as far out as possible to maximise bandwidth and minimise distortion.
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
"The more I read about Stevenson, the more I don't think it applies to modern tonearm. Most of us use 12" LPs almost exclusively and Stevenson doesn't even produce an alignment designed excursively for 12" LPs. "
To say that Stevenson is not relevant to the modern tonearm is to ignore that fact that he is using Lofgren A - it is that you just don't agree with his choice of modulated envelope and recommended universal solutiom! Secondly, he is providing a design guide with recommended offset angles and overhang in relation to the effective length (refer to fig 6 on P.315 of his paper) for anyone not wanting to reinvent the wheel if they intend to design an arm.
He does in fact recommend an alignment suited to 12" LPs - Design 1A. It is essentially the same as the 309 but for the slightly reduced maximum radius of modulation from 146 to 145.3mm and VERY marginal reduction in Xmin from 58 to 57.9mm!!!! The fact that he uses Lofgren A SURELY convinces you that Stevenson is NOT one single set of nulls. I urge you to read the footnotes in the context of the extract shown in this snip.
I have mentioned this before to you, but I disagree with the suggestion that 12" LPs are the only discs that people play. 7" singles and 10" singles as well as maxi 12"s (both 45 and 33rpm) are still being produced and I still continue to purchase them so a Universal solution such as Design 1B is still very relevant for someone wanting to avoid making changes to the aligment everytime a different record is played. I am by no means unique in this respect and there are plenty of inmates who have amassed vast collections of all different record sizes.
Now if you had reworded your statement to be that "no manufacturer has designed an arm that is currently in production that uses the design parameters for Stevenson Design 1B" then I would have no issue at all and would happily replied with a suitable "+1 nt" in the subject line! :)
Now that you have read his paper, there is still one point I raised previously that you haven't addressed yet and that is:
Do you disagree that Stevenson makes the distinction between Xinner and Xmin such that Xmin < Xinner and Xmin != Xinner except in the case of Design 1A where he has determined a modulated envelope where the "typical" minimum radius Xinner coincides with where he requires Xinner to be due to the Lofgren A solution for the envelope with Xmin < Xinner. That will sound a bit circular, but all I'm saying is he is applying Lofgren A to an LP envelope. In this case, he very clearly is not placing Xinner at Xmin as everyone erroneously quotes as a so-called Stevenson alignment.
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
> To say that Stevenson is not relevant to the modern tonearm is to ignore the fact that he is using Lofgren A...
You're right! This is somewhat perplexing, isn't it?
All of Stevenson's null-points are simply Lofgren A null-points when using a modulated groove envelope defined by an innermost modulated groove radius of X 3 and an outermost modulated groove radius of X 2 from Stevenson's Table 2.
So, why do we need Stevenson? Why can't we simply pick the modulated groove envelope of our choice and apply Lofgren's equations? That's what I propose and that's exactly what I've been doing all these years. I just use different numbers than Stevenson suggests. I use the numbers that I believe are best.
Best regards,
John Elison
"So, why do we need Stevenson? Why can't we simply pick the modulated groove envelope of our choice and apply Lofgren's equations? That's what I propose and that's exactly what I've been doing all these years. I just use different numbers than Stevenson suggests. I use the numbers that I believe are best."
That's exactly my feeling as well and how (and why) I created my Quadtractor as I call it. We've been on the same page all along I feel, but arguing on different aspects.
Where we may still disagree is over whether Stevenson actually HAS "an alignment" which can be attributed to his name. I don't believe so given his use of Lofgren A, but I will concede that if Design 1B nulls are considered to be synonymous with his name/paper, then so be it.
What I would still assert is that anyone attempting to implement his "method" should NOT be setting the inner null to match the (absolute) minimum radius of the smallest record they wish to play. This, I feel is where the misunderstanding is over what Stevenson is actually teaching. He is teaching an application of Lofgren equations to achieve minimum weighted peak distortion, so to set inner null = minimum radius is to violate this requirement for minimim weighted distortion for a specified modulated groove envelope.
I think we have successfully reached a concensus over the content/intent of his paper.
Would you agree with the following summary?
i) Stevenson should be viewed simply as a method utilising standard Lofgren equations to optimise alignment according to the record sizes of interest to the end user.
ii) Part 2 of his paper proposes solutions which "teach" someone wishing to optimise their alignment how to choose suitable radii to meet their requirements and provides "standardised" solutions which would suit the typical record sizes for playback as a "one size fits all" alignment. Where a degree of deeper understanding is required, is in understanding which option to choose depending on the predominant record sizes of a particular collection.
iii) Inner null != minimum radius of the smallest record size to be played (or the minimum radius of an LP if that is the only record of interest). The fact that the inner null for Design 1B coincided with the IEC minimum radius is probably where the misconception has originated.
iv) Stevenson teaches a choice of the inner null position based on a "typical" minimum radius with additional consideration to the recorded velocity in relation to the tracking error as a function of groove velocity. He assumed a typical 10cm/s reference peak velocity but acknowledged that peaks of 20cm/s were not inconceivable in the transients of typical recordings especially on 7" singles due to the hotter cut. In other words he is advising the designer to take this into consideration which would influence the choice of inner null position relative to the required modulated groove envelope.
As to whether "Stevenson" is relevant? I would say that would be down to individual choice! For me personally, I think Design 1B is still relevant for anyone who only has 1 deck and needs to play all 3 record sizes and doesn't wish to keep changing alignment.
Thankfully I have 2 decks so I can always have one optimised for 7" and 10" only using Lofgren A alignment with my chosen linear offset of 80.26mm, but for anyone with a single deck, I think Design 1B still has its place.
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
Okay! I think I agree with you although my null-points of choice are still 66-mm and 120.9-mm. I own LPs only and many of mine are audiophile pressigs.
Best regards,
John Elison
Must be Common Knowledge that SME 3009 (Series 2 + 3: early arm wand) are based on Stevenson - the SME protactor (and as issued with S2 headshells) is @ inner 60.325 - matched against my ancient HiFi Choice/ELITE gauge.
I know that 60.325-mm is the inner null-point for Stevenson's alignment, but what about the outer null-point? I can't find the offset for any of the SME tonearms. However, if I could find the pivot-to-spindle distance of the SME Series II, that would tell me if it's possible for it to have Stevenson's alignment with the outer null-point. The parameter that SME states in the manual is "Distance from Bedplate Center to Turntable Center." If this is pivot-to-spindle mounting distance, then it's not possible for SME II and III tonearms to be Stevenson. Sure, you can align the inner null-point to 60.325-mm, but the outer null-point will not be a Stevenson null-point.
What makes you think the SME II and III use Stevenson geometry. If you're going by the null-points published in the Vinyl Engine, those are erroneous. I don't know how the Vinyl Engine came up with those null-points, but they are wrong. Either the null-points are wrong or else the other listed parameters are wrong.
Best regards,
John Elison
SME's old (92mm long) protractor has the 60.325 'null' - which, with the arm aligned to that, seemingly gives an outer 117.42 (SIC!) - when judged against Max Townshend's 175mm long Gauge.. (Personally I use ~Lofgren A as the choice for LP only)Sorry to intrude; but, as UR a previous SME 3 user, thought the above would be Obvious - unless you foreswore larger/more complex OCD gauges 'back in the day'.
Edits: 02/20/17
Hi Frank,
I don't mind the intrusion at all. I would just like to get the bottom of this.
Normally SME provides more parameters than effective length. Nowadays, SME provides pivot-to-spindle and offset angle in addition to effective length. These three parameters define the tonearm geometry completely. In fact, any three independent parameters defines a unique tonearm geometry for which all other parameters can be determined. I truly believe that "Distance from Bedplate Center to Turntable Center" is pivot-to-spindle distance for the specified effective length. Therefore, here are the numbers directly from the manual of the Early SME Series III:
Effective Length = 229-mm
Pivot-to-Spindle = 215.4-mm
.
I owned the "later" version of the SME III, which was designed for Lofgren A geometry within the IEC modulated groove envelope. It had alignment null-points of 66.00-mm and 120.89-mm.
.
I'll take your word for the older Series II tonearm having an outer null-point of 117.42-mm since you measured it. However, if I find any SME published information I'll keep you posted.
Thanks,
John Elison
Just to add (old SME info).3009 templates = spindle/pivot centres: 214mm (original) 217mm (improved) 214.5 (SME 3 - mk.1 supplied with 60.325 protractor, as Series 2).
The later carrying arm (you show) nulls then get re-used (early Series 5 brochure).
The earlier null (can't quote my 1971 protractor..) reflects SME's Classical market - until the mid-60's it would be anticipated to use a mix of LP/45 EP
(NB: THe Series 3 (Mk.1) booklet also quotes 215.4 - but (Steel Rule) the supplied Template is 214.5)
Edits: 02/20/17
> 3009 templates = spindle/pivot centres: 214mm (original) 217mm (improved)We've already addressed the Series III and determined it is not Stevenson's alignment.
Personally, I don't believe the Series II arms are Stevenson's alignment either, but you didn't provide the effective lengths. If you remember from my previous post, we need three independent parameters to define a unique tonearm geometry. For example, we know that Stevenson's alignment includes an inner null-point of 60.325-mm. That's one independent parameter. If we know effective length along with pivot-to-spindle mounting distance, that completes the three independent parameters needed to calculate all other parameters.
Here are the specs from a brochure for the SME 3009 Series II tonearm:
.
Effective length is 9-inches or 228.6-mm. Pivot-to-Spindle distance is 8.43-inches or 214.122-mm. If we adjust offset angle until the inner null-point is 60.325-mm, the outer null-point will be 106.253-mm. That's not Stevenson's alignment. I'm not sure any of the SME tonearm's comply with Stevenson's alignment.
Edits: 02/20/17
I think I know where the problem might be here - bear with me ...the key here is that the SME arms are designed with a fixed linear offset specification and have a fixed angle position for the cartridge. Alignment is achieved by sliding the entire base. The issue here in using the data for effective length in their specs is that there is considerable variability in the cartridge mounting screw centreline to stylus tip (let's call it Tip Distance to reference later) - the main variables of cantilever length and VTA being the determining factors. Typically the tip distance is between 7.5 to 9.5mm although I do have outliers in my collection. An AT440ML is specified at 8.5mm, an AT150MLX is 9mm for example.
Examining the right-angled triangle (for SME) we have already defined the Opposite side (Linear Offset) and the Adjacent (defined in part by Tip Distance in combination with the pipe length). The Hypotenuse (effective length) therefore changes according to the cartridge tip distance since the Linear Offset is predetermined with a specified inner null.
Alignment is achieved by altering arm base which adjusts the overhang and pivot/stylus distance until the stylus hits the correct null point and since the linear offset is defined, you will achieve the second null point where it should be. This is why I think the linear offset jigs are much better because they are completely arm agnostic. You will note that your scan shows "Nominal length" and not Effective Length for the reason I gave about the variable Tip Distance and hence the adjacent. Consequently, the Nominal length should not be used as an input for effective length in your calculations since for a given cartridge other than the one SME designed for, the effective length will always be different.
Would you agree? Or have I forgotten something here!
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
Hi Anthony,
I made my first response way too complex. Here is a simple answer.
Linear offset is fixed as you suggested. Linear offset is defined as the sine of the offset angle times the effective length. However, linear offset also happens to be the average of the alignment null-points. Therefore, when you slide the tonearm base to align your cartridge to the inner null-point of 60.325-mm, the outer null-point must always be 106.253-mm because the average doesn't change. I'm referring to the SME Series II tonearm with its constant linear offset of 83.289-mm.
Best regards,
John Elison
Meanwhile, in the Real World, (devoid of 3 decimal-place precision eyesight) this is what an SME 3009 improved/S2R shell (V15/III squarely mounted/78E stylus) looks like when placed on Stevenson 60/117 pinprick-nulls
The SME 3009/S2 Improved has effective length of 231.2-mm and pivot-to-spindle mounting distance of 215.4-mm. When you adjust the inner null-point to 60.325-mm, the outer null-point will be 116.971-mm. This is Stevenson's alignment. The SME 3009/S2 Improved and 3009 Series II Improved tonearms might be the only SME tonearms with Stevenson's Geometry.
Already stated (20 Feb) the SME 3009-Improved Template measured 217mm - which isn't your 215.4 mm (my quoted 214mm 3009 Template essentially was your imperial) - and SME Series 3 (Mk.1 supplied with a 60.325 protractor) had another published/template discrepancy.Only if you can accurately calculate the distances yourself can these various analyses be provable.
NB: The thicker SME S2-R headshell don't even have a particularly straight edge to align by..
NB2: Decca mastered very infrequently to 58mm - one to hand is the first (1G) transfer of s.1 CS7133/Adler 1978 - also probably applies to Elgar 1/Solti first 1972 issue s.2 (not to hand) - high end-velocities in both instances.
Edits: 02/24/17
> Already stated (20 Feb) the SME 3009-Improved Template measured 217mm - which isn't your 215.4 mmJust because your template measured 217-mm doesn't necessarily mean the pivot-to-spindle distance of your tonearm is not 215.4-mm. You can adjust pivot-to-spindle distance as much as +/-12.5-mm. Furthermore, the exact pivot-to-spindle distance depends on the stylus-to-mounting-hole distance of your particular cartridge. You already know that your alignment null-points are approximately 60-mm and 117-mm and the following alignment equations apply to all pivotal tonearms. Therefore, if you can measure either your effective length or your pivot-to-spindle distance, you can plug the numbers into the appropriate equation to determine the remaining variable.
.
I've already shown that the early SME III with effective length of 229-mm is not Stevenson's geometry. It's null-points are 60.325-mm and 100.188-mm as shown in the graph below. The later version, which is the one I owned, simply had a longer carrying arm with effective length of 233.2-mm yielding null-points of 66.0-mm and 120.9-mm.
Edits: 04/15/17
Have you seen this link before? This gentleman has done a very good job at compiling data for standards and other bits and pieces....
Have a read of the sentence above the table regarding Keith Howard's comment on the SME 309.....and Stevenson 1A! ;)
Imagine how delighted the other inmates are going to be when this is all sorted between us LOL ;)
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
I don't believe I've seen this page before. Thanks for the link!
> Have a read of the sentence above the table regarding Keith Howard's comment on the SME 309.....and Stevenson 1A! ;)
I'm not sure I quite understand the sentence:
"Incidentally this sample linear offset length 91.55mm is near to 91.535mm for SME Series 300 arms (1989) which might be based on Stevenson Type 1A for 12inches LP"
The correct linear offset for the SME 300 Series tonearms is 91.535-mm or 91.536-mm. Keith Howard introduced the linear offset of 91.55-mm only because he rounded off the null-point radii to one decimal place and took their average. He is correct about these null-points being based on Lofgren's "A" alignment for a modulated groove envelope in-between 58.00-mm and 146.00-mm. He is also correct that this alignment is close to Stevenson's 1A alignment. However, it's not identical to Stevenson's 1A whereas it is identical to the Lofgren "A" alignment mentioned above. If it were based on Stevenson's 1A alignment I'm sure SME would have gotten it right and not just close. Anyway, that's my opinion.
> Imagine how delighted the other inmates are going to be when this is all sorted between us LOL ;)
I don't think they really care! If you look at the rest of the page, it appears that life continues without us. ;-)
Best regards,
John Elison
"... He is also correct that this alignment is close to Stevenson's 1A alignment. However, it's not identical to Stevenson's 1A whereas it is identical to the Lofgren "A" alignment mentioned above."
I guess this is really down to whether you consider Stevenson a method or a solution. This isn't something that I would debate, because I think that both views are probably valid in the absence of Stevenson himself weighing in! He would be about 77 now. I wonder if he has internet access? LOL
SME's choice of radii doesn't exactly conform to any of the standards (and Stevenson has also exercised similar freedom with his choice of the maximum radius). However, both are applying Lofgren A to values such that an average user will probably achieve virtually identical solutions when you factor in tolerances and repeatibility.
I view the differences to be analogous to taking PI as 22/7 or 3.14 instead of to say 10 dp in a calculation. You get different numbers, but the solution is for all intents and purposes the same. For me, I am more concerned about the "intent" rather than the actual answer perfectly matching to determine equivalency.
For the vast majority of users who aren't going to fuss over whether the templates are a perfect snug fit over the spindle, and the markings have been verified to sufficient accuracy to a traceable standard, the actual difference in the linear offsets achieved by the majority of users doing things by eye would be inconsequential in that 91.54 or 91.55 are effectively the same thing!
Even with the care I took over constructing my LO jig (I verified the accuracy of the laser etchings to within 10um of the target values), I repeated the alignment process multiple times on a single cartridge to determine repeatibility and verify the design. It took several attempts to iron out the errors due to tiny platter rotations and improve repeatibility from ±0.1mm to within 0.05mm (otherwise it was about ±0.1mm allowing for the slight slop in the pivot reference "pin" which sets the correct arc position for the null). Offset angle is another story. In principle I should be able to achieve ±0.3° but in practical terms, with tapered cantilevers and azimuth errors, it is incredibly hard (in my opinion) to determine when the cantilever is perfectly aligned and I suspect the offset error is larger.... Stevenson himself suggested that ±0.25° within the target value for the offset was the goal, but I really don't think this to be achievable in practice - even with a fixed offset design like the SME, you are screwed if the motor assembly of the cartridge is skewed and a large proportion of my MCs are afflicted (particularly the Denon ones - confirmed by Expert Stylus who had complained to Denon about this issue).
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
> I guess this is really down to whether you consider Stevenson a method or a solution.You've made it perfectly clear to me that Stevenson is the solution to Lofgren's "A" method for calculating alignment null-points from a modulated groove envelope. In fact, you've proven to me that Lofgren "A" is the method used to find Stevenson's null-point solutions.
> This isn't something that I would debate, because I think that both views are probably valid in the absence of Stevenson himself weighing in!
Your logic appalls me! Firstly, both views cannot possibly be valid. Secondly, "Stevenson weighing in" cannot possibly resolve anything. The only person that can resolve this issue by "weighing in" is SME.
> SME's choice of radii doesn't exactly conform to any of the standards (and Stevenson has also exercised similar freedom
> with his choice of the maximum radius). However, both are applying Lofgren A to values such that an
> average user will probably achieve virtually identical solutions when you factor in tolerances and repeatibility.Tolerances have nothing to do with this discussion. We are talking about the basis for a specific tonearm geometry---not how accurately a user can achieve an intended alignment.
> I am more concerned about the "intent" rather than the actual answer perfectly matching to determine equivalency.
The only way we have to determine "intent" is to analyze the numbers. I showed you that Lofgren's "A" solution for an innermost and outermost modulated groove of 58.00-mm and 146.00 produces SME's parameters for the 300 Series exactly out to two decimal places. In this respect, Stevenson's number don't even come close.
Parameters ..................... SME .............. Lofgren ................. Stevenson
Effective Length ........... 232.32 ............ 232.320 .................. 232.320
Mounting Distance ....... 215.35 ............ 215.345 .................. 215.448
Offset Angle ................. 23.204 .............. 23.204 .................... 23.124You're off on a tangent now with all your gobbledygook about alignment accuracy. It doesn't pertain to this discussion. The SME Series 300 tonearm parameters are clearly specified by SME and they will not change regardless of whether you and I believe they are influenced by Stevenson or Lofgren. Therefore, it's really not worth arguing anymore, is it?
I will say I've had fun with this discussion, though. I enjoy our interactions in the Asylum. Thank you very much!
Best regards,
John Elison
Edits: 02/25/17 02/25/17
Come come "Appalling logic" and "Gobbledygook" are a little strong John! I know I have difficulty expressing myself clearly at times and VERY occasionally get sidetracked... ;)
Anyway, I agree, we've battered the hell out that topic! I had fun too and I think we had a very productive exchange which will hopefully be seen one day through the archives for anyone questioning the same topic. I always enjoy a good mental "sparring" session like this with you and I thank you as well.
BTW, I don't know if you read this Keith Howard article in Stereophile, but I think you would find this a very interesting read and is right up your/my alley
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
> Come come "Appalling logic" and "Gobbledygook" are a little strong John!
You're right! I apologize! I suppose that both views could possibly be valid, but it would require SME to "weigh in" to verify that conclusion with regard to their 300 Series tonearms. With regard to accuracy in the alignment process, what you say is valid; it's just irrelevant to resolving the tonearm geometry issue of the SME 300 Series tonearms.
Thank you for introducing me to Kieth Howard's article on pivotal tonearm geometry. I briefly skimmed it and I will go back and scrutinize it thoroughly when time permits. He makes some very interesting points. Of course, I think his argument for shorter tonearms versus longer tonearms is highly exaggerated. I wrote a post on this very subject some years back. After all, an air-bearing linear tracking tonearm in nothing more than a pivotal tonearm with infinite length. I don't thin anyone would argue in favor of a 9" tonearm versus a linear tracking tonearm on the basis of alignment error.
Another thing Keith forgot to mention is that SME tonearms can be aligned with a one-point protractor as accurately or even more accurately than with any other type of protractor. That's the beauty of the SME design.
Anyway, I have issues with some of his other points, but I think it would be better to leave them for a different thread and a different time. However, it might be fun to discuss this article further when we get a chance.
Interestingly, I met Graeme Dennes in person several years ago and he is the one who convinced me to begin promoting Lofgren as the father of modern tonearm alignment instead of Baerwald. We exchanged many emails discussing tonearm alignment and other audio related topics. I wrote the post at the link below after meeting with Graeme Dennes.
Thanks again,
John Elison
No, I don't agree! However, it's not me that doesn't agree; it's tonearm geometry that doesn't agree. That's the beauty of linear offset. A specific linear offset will always produce the same outer null-point regardless of the stylus-to-mounting hole distance. That's also the beauty of the sliding base principle of alignment. You don't need an arc protractor or even a two point protractor because the outer null-point will always be the same after aligning the inner null-point. SME has chosen the very best method of tonearm alignment. Let me show you.
Here is the alignment of the SME Series II tonearm I presented in my previous post.
.
Now, suppose we use a cartridge with stylus-to-mounting hole distance that produces an effective length that is 5-mm longer than spec. We will still use a protractor that sets the inner null-point to 60.325-mm by sliding the base farther away from the spindle. The cartridge remains squarely mounted in the headshell and the linear offset remains the same.
.
Now, let's use a cartridge with stylus-to-mounting hole distance that produces an effective length 5-mm shorter than the original. We now slide the base closer to the spindle to align our cartridge to 60.325-mm inner null-point. The outer null-point remains the same as before because the linear offset has not changed.
Forgive me for saying this, but I'm not sure you are reading my (admittedly wordy) messages in detail as you appear to be repeating back the same information I wrote in my original message as though I need an explanation! I believe I had already covered the advantage of linear offset based jigs many times before and how the arm base slide is adjusting the overhang to the new effective length defined by the cartridge in use. Would I be wrong in interpreting this as a retreat from your previous advocation for arc protractors? The archives will show that I have repeatedly stated that Linear Offset jigs were arm agnostic and able to achieve higher accuracy in alignment.
In fact I would argue that the use of a Linear Offset jig with an arm utilising headshell slots combined with interchangeable headshells gives an even more versatile and accurate solution than SME - recall I made a jig that provides me with 4 different alignments which I can swap between at will with a simple change of the headshell mounted cartridge and achieve whatever alignment I deem to be most appropriate at the time!
Anyway back to your post - I tried to point this out gently in my previous message, but I'm afraid your linear offset calculation is incorrect because you are taking the "nominal length" as an actual specification. Your calculated offset angle is too low for the length of the arm; There is no way 83.29mm would be the correct linear offset. Your previous screen shot clearly indicated that SME have designed the arm for an inner null of 60.325mm (2.375"). Anything less than 88.65mm is no longer applicable to an LP as it puts the nulls too close together so they no longer satisfy the minimum peak weighted distortion solution of Lofgren A for typical LP radii unless you dramatically reduce the modulation envelope to an Xmax of <126mm or so with Xmin of somewhere around 56mm. You have previously stated that SME do not make mistakes and I am inclined to agree (although the 2.75" inner null for the 3009 Series 1 would perhaps suggest otherwise and they made subsequent corrections).
Frank's information on the null points he achieves with said arm would tend to confirm my view that there is a mistake in your calculation of the linear offset.
Sorry, I must respectfully disagree with your assertions based on your calculations as they are inconsistent with the evidence and data available.
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
I'm sorry! I probably do not understand everything you're writing. My mind is a bit foggy with all the pain medication I'm taking currently.
It just dawned on me what you mean by "linear offset protractor." It's the type of protractor that you point at the tonearm pivot like a Dennesen Soundtractor. Yes, it appears that you have constructed a very nice one. I have no doubt it is accurate.
On the other hand, I prefer an arc protractor for tonearms with fixed mounting distance and variable effective length. For tonearms like SME with variable mounting distance and fixed effective length, a single point protractor is all that is necessary for utmost accuracy. In other words, I'm sure your linear offset protractor is just as accurate as any arc protractor, but I still prefer an arc protractor for tonearms other than SME.
> I'm afraid your linear offset calculation is incorrect because you are taking the "nominal length" as an actual specification.
I believe the "nominal length" is an actual specification for a cartridge with the right stylus-to-mounting-hole distance. In that case, I believe the two lengths provided represent effective length and pivot-to-spindle distance. If these numbers do not represent those two independent parameters, then you might be right about my calculations. If you can find any SME literature that provides three independent tonearm parameters, I would be very interested in seeing it.
I apologize, but I'm too tired to continue tonight. Please show me any SME data you have to support your position. If you've already done this, then I apologize again. I'm just too foggy at the moment to discuss this further.
Thanks,
John Elison
Thank you - you beat me to it!
Regards Anthony
"Beauty is Truth, Truth Beauty.." Keats
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: