Localization is not only like owl's pouncing on mice at night, it's to give separation to the instrument layers, and for indoor recording, perceiving the tone color the hall contributes to each directional reflection, summed at each listener's individual pinna to create an "extension" of each instrument the musician plays differently in each room depending on its acoustics. Standard orchestral micing captures a recording angle of 120deg, spatially compressed on playback by your speaker angle, say by half to 60deg. Squeezing to 60deg stereo, the distinguishing of auditory events is critically affected. Of course separating your speakers 120deg would make for an intolerable hole in the middle. (An Ambiophonic setup reproduces the original 120deg with no hole in the middle, and no comb filtering of soloists in the center.)
Even without a $25 capacitance meter (Circuit Specialists, Tempe AZ) to label your phono interconnects, you might estimate your cartridge's capacitive loading: e.g. 125pf for 5ft of 25pF/ft cable plus 150pF typically inside the preamp, for a total 275pF many cartridges specify. Manufacturers of MM\MI pickups intended their products' equivalent circuit be completed by the user's load capacitance & resistance, 47kR by convention. Yet magazine reviews continually criticized a pickup as "too bright" or "too dull" when this was probably due to their own negligence.
Please refer to my reply to PANO above re mixing phono signals to mono.
I like your listening room. One of mine of yore had a Hammond H100 and Leslie, plus a Steinway "O" grand.
Re experiencing 3D, if you were 40mi from me I'd pick you up!
Last edited:
Agree, PANO. Most stringent among recording for musical genre for the "you are there" illusion, the conditions for being able to localize an orchestra section if not individual instruments soling back in those sections are: 1) a properly made spatial recording in an acoustically good hall or large studio capturing both ILD & ITD using a main near-coincident mic pair or surround array (maybe a few spots lower) but not wide-spaced (having little ILD but too much ITD time delay whereby voices jump around) nor coincident (having only ILD but lacking ITD); 2) reproduction attending to speaker dispersion and controlling listening room acoustics. Of course easier "they are here" pop music reproduction will benefit as well.
I think this deserves it's own thread for looking at the methods. One of the things on my ever growing list. I recently had one of my ebay purchase MCs converted to mono by rotating the generator 45 degrees. Now right channel is sum and left is difference. Now the challenge is to compare that with other methods of getting to the same end. I think series strapping should perform pretty well.
Last edited:
Yeah, I'd like a thread like that. 
I have found that a great deal of the noise on my scratchy old mono LPs is out of phase or unique to one channel. Lose that and the noise goes waaaay down.
I have found that a great deal of the noise on my scratchy old mono LPs is out of phase or unique to one channel. Lose that and the noise goes waaaay down.
Based on some research done on here I am not sure I believe manufacturers loading specs. The MM generator is a relatively complex impedance. I honestly think Barney Oliver was onto something 50 years ago which was widely ignored on a better way to do things.
Yeah, I'd like a thread like that.
Is that an artifact of using a stereo cartridge to reproduce mono?
Yes it is. However I consider it a useful artifact if I want to record vinyl. Makes it easy to get rid of so much that's not meant to be there. Not in both channels? You don't belong!
Scratches and dirt, being ruts and boulders to the stylus, mostly force it up & down. Vertically, stereo cartridge signals are opposite polarity, so summing them for mono reproduction cancels much of the noise. But maximum cancellation does not happen unless the signals are first balanced. Most cartridges have some imbalance in sensitivity, typically specified as 2dB max. (Most tape machines have individual channel level controls because head stacks have similar manufacturing variances resulting in difference in channel sensitivities.) If L & R are different by 2dB, then vertical noise is reduced only ~12dB when balance has not been restored in the preamp before mixing L+R, also when cartridge channels are simply shorted. As a 10dB reduction is perceived as about half volume, a partial 12dB reduction in noise falls short of the audiophile objective. My go-to Stanton 881Sii's measured imbalance happens to be 2dB, so preamp's I design & build such as the project in my book have channel gain controls for balancing the signals and then properly mixing them to mono, which again cannot be done by simply shorting channels without loading distortion.
MM\MI pickups loaded incorrectly by users-supplied load resistance and capacitance dramatically affects frequency response, with too low capacitance causing a broad dip in mid frequencies and a troublesome ultrasonic peak. Too high a load capacitance does the opposite, a broad rise in mid frequencies and rapid falloff in highs. This could explain magazine reviewers' erroneous criticisms of a cartridge sounding either "too dull" or "too bright." If a user's C-loading is correct, the cartridge will meet its frequency response spec and sound as good as it can.
Reputable manufacturers implies they publish meaningful specs. For recommended load capacitance, specs can range from 100 to 675pF. On the low end, and using 5ft of good 20~25pF/ft coax resulting in 100~125pF, the preamp should add none. For higher, it's easier to add capacitance than to reduce it. For many pickups, a reasonable value of 275pF is typical, so preamp's often have a 150pF capacitor at their input jacks, adding to the cable's 100~125 to get within the required 10~20% of spec.
Last edited:
Have you ever actually measured generator performance over frequency? The inductance varies over frequency. Getting a decent electrical model of the generator (done by some members here) shows some very interesting things.
As you missed the comment on Barney you might have some research to do...
As you missed the comment on Barney you might have some research to do...
That's about where my Stanton 681 is. I thought the imbalance was near 2dB, but with more careful wiring and measurement, found that left and right are very close.
Why do you suspect (and express rudely) that I've not measured the performance of cartridges? The best of those I've measured is +\-1/2dB 30~15kHz including the CBS pink noise test disk and thus within its limits. I've also measured extensively MM\MI frequency responses due to capacitive loading, styus tip dimensions and evaluated wear under a 400x microscope, optimized cantilever-tonearm resonance, modeled & simulated cartridge equivalent circuits, measured the performance of many phono preamp's of my and others design, and published all these along with the science behind grooved media in my book, which your rude comment proves you have "missed." I didn't miss your "Barney" comment, if you'd please provide a link, as I have a long list of research to do.
Last edited:
Compared to other anomalies, any cartridge channel sensitivity imbalance is an almost unavoidable result of variations during manufacturing. Also no big deal, as it can easily be compensated. What matters more is that the channels match in other ways, such as frequency response, which your 681 was certified for in writing by Stanton. Mastering engineers used the 881 for QA during lathing, adjusting for those measurements for this MM. My 881Sii has the full 2dB imbalance, but the channels are precisely matched in every other way, both measuring and listening. So I wouldn't worry whether your 681 has 2dB or 1/2dB channel-to-channel balance so long as your preamp has separate channel gain calibration.
Microphone preamp's all have gain controls for the same reason: I've published that mic mfgrs charge a premium for a pair that typically they only match in sensitivity, while I purchase them on the basis of holding a pair of FR printouts up to the light in order to match them by similarity of FR regardless of output level. Otherwise, voices jump around the recorded soundstage from bogus instantaneous ILD changes by pitch as they're playing. And that same thing happens when any link in your audio chain isn't matched in frequency response with its counterpart in the other channel.
Last edited:
I would put forward that you are the rude one by not looking at work done on here. Scott for instance has actually compared pink noise sweeps across different test records. The CBS does not come out top! Attached is cartridge equivalent circuit Hans Polak generated from a sweep up to 1MHz. That is more accurate than anything I have seen published to date. There has been a lot of research done on here. As you are keeping your cards close to your chest you maybe can add to this, or are possibly some years behind where DIYAudio is.
I am not sure if you are here to lecture or to learn. Hopefully the latter...
I am not sure if you are here to lecture or to learn. Hopefully the latter...
Attachments
Hello,
can you provide approximate wear of stylus and grooves after how many plays ? There was thread on here but I forgot. And would be nice to see extra study results.
Thanks and regards.
OK, I revisited the test disk thread* of 2016 - four years after 2012 when I began assembling my book (published 2017 and updated online). Not that he'd want to get dragged into this, I see Scott's mentioning "RCA's STR-101" (actually CBS's STR-101) which, being CBS Labs' first, has long been replaced by the STR-150. I use most of the CBS series; the pink sweep of the STR-140 that I prefer minimizes replay resonance issues of sine sweeps, and when properly interpreted is "tops" in the estimation of my peers (fellow audio engineer members and colleagues of AES, SMPTE, ASA, CAA, and VDT). This 2-page thread is not a rigorous treatment - perhaps there are others? Too bad the group's test disk design has not been realized.**
Since I joined DIYaudio in 2007, 13 years ago this month, I've been a member both to learn and to share (within the space limitations of a forum post). Please consider that this comes from one who writes scientific papers and books. I will appreciate thoughtful constructive criticism, accompanied by specific references, and sticking to the point. But if I inadvertently merely offend your ego, please just ignore me.
*calibrating test records
**https://www.diyaudio.com/forums/analogue-source/313335-test-lp-buy-8.html#post5211843
Last edited:
Hiten, you ask a simple question, but unfortunately I can't think of a simple answer. Of the many styli I've observed and photographed, some lives are cut short by the bonded tip falling off, bent\broken cantilevers, or improper\no anti-skating, which can wreck one side of the tip long before the other. These are the worst cases in my experience. For more typical cases, measuring stylus wear is not precise, and your mileage will have many variables, so I'll give only round numbers.
And hours aren't the only factor. You know that wear varies somewhat linearly tracking pressure (mildest wear of both the tip and grooves <2g; accelerating >3g). OTOH highly variable, dirt in the groove is a pumice like jeweler's rouge. Also that the effect of skating and its compensation is only very approximate, as each disk can have vinyl with a different coefficient of friction.
Ellipticals wear sooner (some <500hr) than sphericals, but light-tracking line contacts wear longest (1,000hr or more - 2,000 sides!) yet create the lowest tracing distortion. I have seen Stereohedrons with an estimated 400hr tracking at 1.25g that under the 'scope look almost new - they will likely outlast me! The tell-tale is a just visible flat that will develop by 800~1,000hr to have a sharp shaving edge around it. [Several score microphoto examples and a wear v. VTF v. HF cutoff tables are in my Phonograph book.]
A few more anecdotes: Most demanding, a radio station client's three turntables had sphericals which I estimated had 1500hr on one side and much worse than I've ever seen on the other, so had been destroying their 8,000 disk library. I replaced them with ellipticals in the on-air studio and a line contact in production for highest quality transfers. Disks themselves exhibit a bit of high frequency erasure with only the first few plays. Spherical styli can ride better over embedded dirt without reproducing the boulders and scratches as loud clicks.
There's a lot more, but I hope this is helpful start. After 59yr as a professional audio engineer, I'm seeing more every day. - Robin
Last edited:
I think Scott's Knowledge is nearly impossible to beat. Experience endless.
However, RCA guy brings new life into themata, but until now no real news from my view.
Where he is right, that most distorstions are trackingdependent.
Cutting lathes work different than any tracker stylus due geometrical layout of the cartridge/ Cutting head
Va bene.
Conical stylus introduces almost K2 , other shapes K3. Both with overtones.
Fact is, both styli do not track exactly what is recorded. Mission impossible.
Angle errors.
With perfect flat and round records, the tracking angles would be within more or less 1 degree. Resonances and other bugs increase it up to maybe 5 degrees.
Now the the 1 Million$ Question, is it better to have K2 or K3 plus Spectrum.
Dynagroove etc. addressed the problem with K 2 and controllable to some degree and did predistorsion, with K3 and up no chance for predistorsions.
Furthermore, Rumble in the Jungle over /under 10-11 Hz creates new hard to fix bugs for ported speakers. Closed boxes or and soem other systems are pretty immune here.
I use a system, suppressing anything below 18Hz with 60 dB/ Octave, but still long excursions visible at woofers . Audible not really, but sometimes i feel the pressure of the air somehow.
What RCA guy says abouth capacity and so on is ok, but nothing new.
Lets see what's going on.
"(...) pre-echo of the loud sound occurring 1.8 seconds ahead of time."
Out of curiosity, I've heard a few records that had reproducible pre-echoes right at the beginning of the first track (usually rock music). But these pre-echoes seemed to be much less than 1.8 s early, more like 300 ms or so. Back then I thought the master tape had been re-used and not erased properly, but that was likely way wrong. The short pre-echoes were reproducible, except very quiet so only detectable when using headphones. I just don't remember which LPs they were, gives me an excuse to go through the collection with my new headphone amp!
--Christian
EDIT: maybe they were 1.8 s ahead after all...
Out of curiosity, I've heard a few records that had reproducible pre-echoes right at the beginning of the first track (usually rock music). But these pre-echoes seemed to be much less than 1.8 s early, more like 300 ms or so. Back then I thought the master tape had been re-used and not erased properly, but that was likely way wrong. The short pre-echoes were reproducible, except very quiet so only detectable when using headphones. I just don't remember which LPs they were, gives me an excuse to go through the collection with my new headphone amp!
--Christian
EDIT: maybe they were 1.8 s ahead after all...
Last edited: