Audio Asylum Thread Printer
Get a view of an entire thread on one page
|For Sale Ads|
In Reply to: RE: Null points for 7" discs? posted by flood2 on July 05, 2017 at 17:03:29
You might want to recheck your distortion figures. I get different numbers using a 230-mm effective length tonearm.
Remember, the modulated groove envelope is in-between groove radii of 53.975-mm and 84.1375-mm. With my null-points of 56.97-mm and 77.77-mm, I calculate the following:
Average RMS Distortion = 0.108%
Peak Distortion = 0.159%
With your null-points of 53.975-mm and 76.23-mm, I calculate:
Average RMS Distortion = 0.137%
Peak Distortion in-between null-points = 0.195%
Peak Distortion at outermost groove = 0.218%
Peak Distortion at innermost groove = 0%
It seems that your alignment produces higher levels of distortion everywhere within the modulated groove envelope except for the innermost groove where your inner null-point is coincident with the innermost groove.
Um, I think you might want to check your assumptions! ;)
For instance why are you assuming a fixed effective length tonearm?? It isn't applicable in the scenario suggested by the OP:
The OP is using an arm with a replaceable headshell so the fixed parameter will be the pivot/spindle distance. For want of a suitable spec, I chose 215mm assuming a typical 9" arm such as a Technics arm. For convenience, I just used the VE calculator which will give slight errors to your calculator due to rounding. However, the figures are correct using these constraints.
Secondly, you have simply chosen to compare the two over the reduced modulation envelope, whereas my frame of reference was with respect to the slightly wider envelope so naturally you will see a bias in favour of the theoretical RIAA/IEC optimisation. If you compare the two geometries using these two constraints, you find that my suggested nulls are superior for radii < ~55.5mm and will continue to be superior for any 7"s that are cut further in than the standard.
As I pointed out, the distortion figures are based on a simplified model which even for an LP suggest that Lofgren A over the IEC/RIAA envelope doesn't necessarily give the best result on the inner grooves. I have a 12" 33rpm cut at an insane amplitude up to ~52/53mm and several more cut well beyond 57.5mm with some around the 54mm region.
When you are talking about a 7" which may run at 33rpm, then I believe the effect of the groove velocity becomes ever more important particularly when the radius of curvature of the tip is becoming a signficant parameter with respect to the wavelength of the cut signal at HF. Consequently, as the rate of angular error with respect to radius increases, the distortion will increase at a higher rate than suggested by the standard model. I don't know what stylus design the OP is using, but certainly improvements are audible in the imaging with a MicroLine or similar type of Line Contact that retains full bandwidth to the innermost grooves.
I therefore assert that a more optimal solution is to be had by accepting a marginally higher tracking error at the outer edge and setting Rmin at a radius somewhat less than the minimum groove on the record which will give a gradually reducing peak weighted distortion envelope. In fact Stevenson was alluding to this with his solutions. You will notice that not only had he chosen to apply Lofgren A to smaller Rmin values, but he also chose reduced Rmax values. When you analyse his solutions over the wider envelope specified in the standards, you do indeed achieve a gradually decreasing peak weighted distortion envelope with respect to radius.
BTW I am extremely impressed if you can hear an absolute increase of 0.036% distortion between the nulls, although I'm sure you would point out that this translates to an increase of 22.6%..... :)
"Beauty is Truth, Truth Beauty.." Keats
> For want of a suitable spec, I chose 215mm assuming a typical 9" arm such as a Technics arm.
Okay! That means effective length will be 225-mm instead of 230-mm. It increases my numbers only very slightly. If you align your Technics SL-1200 for a 7" record, it will have an effective length of 225.067-mm with an offset angle of 17.417-degrees. Here are the other specifications:
Innermost modulated groove radius = 53.975-mm
Outermost modulated groove radius = 84.1375-mm
Inner null-point = 56.966-mm
Outer null-point = 77.773-mm
RMS distortion within the modulated groove envelope = 0.110305%
Peak distortion in-between the two null-points = 0.162853%
Peak distortion at the outermost groove = 0.163073%
Peak distortion at the innermost groove = 0.163029%
The argument in favor of your universal alignment is not applicable here. Damian asked for an alignment optimized for 7" records and that's what I gave him. You can blow smoke all you want, but you can't have it both ways. All your previous arguments addressed the fact that your universal alignment was designed for different sized records. The alignment presented in the graph below is optimized for 7" records and it provides the lowest tracking error distortion possible in-between the green lines, which represent the modulated groove envelope for a 7" record.
"The argument in favor of your universal alignment is not applicable here."
Hey John, the 60s just called....they want their BONG back!
Either you are TUI (Typing Under The Influence) or you really seem to not read carefully these days - I've never mentioned a Universal Alignment in this thread so I have no idea what you are on about. LOL!
"Damian asked for an alignment optimized for 7" records and that's what I gave him. "
Sorry to disappoint you, but there are no prizes for coming in second.. 8-D
Do you own any 7"s? I suspect not, because if you measure them, you'd find most start around ~83mm to 83.5mm to give an increased margin for handling and cueing.
The "correct" optimisation is therefore the one that matches to the records you want to play not what the standard defines (which of course is a reasonable starting point in the absence of any specific parameters)
The OP has said he has many headshells so he can set up a cartridge for each set and tell us what he finds! I suspect he won't hit either solution accurately enough to be able to tell the difference in a blind test.
Let's be clear here - the Standards are a guideline but are not mandatory. Cutting engineers can and obviously do routinely deviate since the envelope is a balance between playing time and "loudness". Since the 80s, loudness is the first priority for radio play.
45 rpm singles are cut much louder than EPs and using my collection as an example, it is evident that Rmin is typically ignored in order to satisfy the requirement for a reasonable playing time given the loudness constraint. A significant proportion of my 7" 45s are cut in to 53mm and the majority at start around 83(ish)mm. The OP is likely to have similar starting points and end points.
If you want to optimise for a 7" then you should be optimising for the statistically most likely envelope which puts it at around 53 to about 83.5. I added in a further optimisation to accommodate the rapid increase in tip related distortion (which you obviously haven't considered) and set Rmin at 51 to because outliers also exist, giving a Linear Offset of 65.1 compared to 67.3 and this gives a superior optimisation for R <55.5mm when compared to yours. Even if he DID have a record starting at 84.1mm, the OP won't hear the miniscule penalty in distortion which is already going to be dwarfed by the transducer and tracking related distortions. He will in theory, get a better result at the end of a side.
As usual you are splitting hairs within hairs in order to try and prove a point. Sorry to be blunt, but you just blindly quoted the standard and ran to your favourite calculator to impress everyone (else) with 6 decimal places and generate null points to 3 dp giving completely unobtainable solutions (in terms of accuracy) and claiming it to be the correct solution.
The Best solution is the one that works best for the end user. I don't care if he uses my suggestion or not. However, I know that it WILL give him superior results in his application with the experience and knowledge to back it up.
Oh, and just for the record (pun intended)....My house is Smoke Free! ;)
"Beauty is Truth, Truth Beauty.." Keats
> As usual you are splitting hairs within hairs in order to try and prove a point. Sorry to be blunt, but you just blindly quoted the standard and ran to your
> favorite calculator to impress everyone. The "correct" optimization is therefore the one that matches to the records you want to play not what the standard defines
I believe you are the only one splitting hairs. If you took the time to read my original answer to Damian , you would know that I provided all the tools necessary for him to choose any modulated groove envelope he wanted and calculate the optimal set of null-points for his own record collection. You, on the other hand, just keep blowing more and more smoke.
> Let's be clear here - the Standards are a guideline but are not mandatory.
You are very wrong. Standards are mandatory; otherwise, you end up with unplayable records. It would be absurd to set up your turntable for outliers and records that deviate from the standards unless you intend to readjust your tonearm for each individual record. Outliers and records that deviate from the standards are exceptions to the rule. You want your turntable optimized for the vast majority of records, not the outliers.
> The Best solution is the one that works best for the end user.
I know! That's why I'm wondering why you're arguing with me. I'm not the end user. Just like you, I presented my case and provided the end user with the tools necessary to solve his problems. I didn't attack you or your answers.
I'm sure Damian can make up his own mind how to align his tonearm. Therefore, you don't have to try to convince me of anything else.
Thanks, but no thanks!
Lofgren "A" Tonearm Alignment for 7-inch Records
Flood2's Recommended Alignment for 7-inch Records
"You want your turntable optimized for the vast majority of records, not the outliers"
That's what I said..... look at the JIS standard...53mm for innermost radius for a 7". That would be an outlier with respect to your RIAA reference...but not to the JIS standard. As it happens, a large proportion of my 7"s are cut loud and go in to 53mm. They are cut in the UK, Germany, US, NZ, Australia and Japan. None start at 84.1mm but 83 to 83mm.
You are missing the point as usual about your assumptions on the distortion calculation. They are oversimplified and you need to consider the higher groove modulation for 45rpm singles. Consequently it is of greater importance to minimise the distortion at the innermost grooves as the increase in distortion due to amplitude and radius of curvature is greater than your calculator predicts. This requires an adjustment to the inner null which necessarily involves a reduction in the Rmin value used for Lofgren A.
Also, I take exception to your labelling of the second graph as "Flood's Universal Tonearm Alignment" - you are deliberately misrepresenting me. It ISN'T a universal alignment. For you to say that implies you don't understand what I am saying. You can't seem to get your mind off it for some reason.
It is Lofgren A applied to a modified envelope that still produces 2 null points on a 7" record.
Please correct that.
"Beauty is Truth, Truth Beauty.." Keats
> Consequently it is of greater importance to minimize the distortion at the innermost grooves
You used the plural form of "grooves" meaning multiple grooves. In other words, you seem to be talking about minimizing distortion across the inner groove area and not just at the innermost groove. You can accomplish this more effectively by placing the inner null-point in the middle of the inner groove area rather than at either end of it. This is what Lofgren's alignment does. This is why Lofgren's "A" alignment has the lowest distortion possible in-between a given modulated groove envelope whereas your alignment with the inner null-point placed exactly on the innermost groove will have greater RMS distortion throughout the inner groove area as well as throughout the entire modulated groove envelope.
On the other hand, it probably doesn't make that much difference since the modulated groove area on a 7-inch record is so small anyway. When you compare the distortion curves for a 12-inch record with a 7-inch record, you can see that a 7-inch record has much lower tracking error distortion as result of its smaller size. However, Lofgren's "A" solution for a 7-inch record still has slightly lower distortion than your solution.
Lofgren's "A" Tonearm Alignment for a 12-inch Record
Lofgren's "A" Tonearm Alignment for a 7-inch Record
Flood2's Recommended Alignment for a 7-inch Record
I think you will find the link interesting - this gentleman has done a very detailed analysis of the different standards and the screenshot comes from his page - I thought I had sent it to you during one of our previous discussions, but I can't remember.
"You can accomplish this more effectively by placing the inner null-point in the middle of the inner groove area rather than at either end of it. This is what Lofgren's alignment does."
This is exactly what I did! ;) Just to clarify, we don't disagree on LofgrenA as a method. We disagree on the Rmin and Rmax values to calculate a solution.
Your insistence that I am choosing the "innermost" radius (Rmin) = inner null (which used to be your original interpretation of Stevenson until we thrashed that one out) is predicated on an assumption that the 54mm Rmin value in the RIAA standard is to be considered a hard limit whereas it seems more of a recommended value which would be considered "good practice" for a cutting engineer to work to as opposed to a specific engineering requirement which would directly affect interchangeability/functionality which would otherwise make it a mandatory specification. The JIS standard as I pointed out specifies Rmin to be 53mm therefore the 54mm in the RIAA standard cannot be considered a hard limit and it is clearly not based on an engineering constraint. Just like for LPs, DIN and JIS specify much small radii than the 60.3mm you base your calculations on.
The clauses in the standard governing the number of turns possible in the lead out before reaching the label clearance zone and the finishing groove (which for a 7" is between 48.2 and 49.2mm) are the key and these therefore give rise to an uncertainty in the exact Rmin value to take. You will find that the Rmin value specified by Aardvark would be "the accepted value" which is intermediate to a range that meets all the requirements in the engineering specification. This would come down to cutting facility policy regarding the preferred clearance area in order to mark matrix codes, add funny messages or for the cutting engineer to sign his name! I don't believe it has anything to do with tracking error related distortion. If this were the case, then the novelty CD I saw on Youtube (look up Techmoan!) which had a playable surface for a turntable on the label side would be unsellable if compliance to the standard was mandatory. We aren't talking about electrical safety or medical devices here in which case the standard ARE mandatory!!
"...doesn't make that much difference since the modulated groove area on a 7-inch record is so small anyway"
I'm glad you finally agree on that! As I pointed out ages ago, the difference in our solutions was miniscule so our disagreement really only comes down to what value to take for Rmin.
So the optimal solution (as I stated previously) which we both agree on, is the one that matches an envelope that one is statistically most likely to encounter.
For exactly the reasons you point out, it would be foolish/inadvisable to set the inner null at Rmin (the extreme outlier value) and this is NOT what I have been advocating.
What I have been driving at from the beginning is that the calculations for distortion are over simplified (assuming a 10cm/s maximum velocity) and do not adequately model the rapid increase in the end of side distortion. Past the inner null, d(Angular error)/d(Radius) is an ever increasing value. However, the sensitivity of the distortion to small errors in SRA (for example) suggests that the cumulative level of distortion increases at a far greater rate. If the gradient were constrained within limits (as between the nulls), the distortion is less offensive unless the amplitude of the groove is sufficiently large in which case I can clearly hear changes in the coarsening of the sound across the record and between the nulls.
The resultant distortion is very tip profile alignment and design sensitive as the radius of curvature with respect to the groove wavelength becomes of a similar dimension. Therefore further gains are to be had (which I believe I achieve in practice with my modified Rmin and Rmax values... although you are perfectly at liberty to dismiss this as wishful thinking!) by biasing the choice of nulls/linear offset to favour a gradually reducing peak weighted distortion across the side. This is achieved by applying the standard Lofgren A equations to a slightly reduced Rmin and Rmax. As it happens Stevenson 1A does just that and he chose Rmax 145.3 and Rmin 57.9. When you apply this to your preferred envelope from 146.3 to 60.3 you get a gradually decreasing peak weighted distortion with a very small penalty in the maximum distortion at 146.3. However, the high groove velocity makes this largely imperceptible.
Stevenson 2 does the same for a 7" and my experiments with that alignment gave a preferable result to the standard application of LofgrenA to the "standard RIAA" envelope.
"Beauty is Truth, Truth Beauty.." Keats
We're good! Or at least I think we are ;)
This is just "creative dissent" (mental sparring) and John and I do this from time to time...
In all seriousness, the take home point for you is to choose the modulated envelope that you find most common in your collection.
Use John's formulae to calculate your null points. In the absence of any measurements, by all means go for John's recommendation.
My collection has a significant proportion cut further in and LOUD so my optimisation takes that into account and it is worth pointing out that the RIAA standard is not the only standard (there are country specific standards) and the minimum radius specified in John's optimisation is not an absolute hard limit based on an actual engineering constraint for functionality. In a standard, the wording is critical in determining a specific requirement vs a guideline. In some cases, the clauses are open to interpretation - the minimum radius being an example.
"Beauty is Truth, Truth Beauty.." Keats
We just have a difference of opinion regarding tonearm alignment. ;-)
Post a Message!
This post is made possible by the generous support of people like you and our sponsors: