Are the T/S parameters provided by manufacturers:
1) values for fresh, unused drivers?
OR
2) values for broken-in drivers?
1) values for fresh, unused drivers?
OR
2) values for broken-in drivers?
For broken-in, after 2-hour "massage" with pink noise at high enough power. That is enough, believe me, not the often quotes of some "north of 100-hour".
But do not believe in manufacturers data, some are spot-on, some are way off, it is best to measure them yourself.
But do not believe in manufacturers data, some are spot-on, some are way off, it is best to measure them yourself.
Thanks 🙂
I ONLY want to know whether the data provided by manufactures is supposed to be for fresh or broken-in drivers.
Whether, or not, to believe manufacturer's data is a different topic and not the subject of this thread.
I ONLY want to know whether the data provided by manufactures is supposed to be for fresh or broken-in drivers.
Whether, or not, to believe manufacturer's data is a different topic and not the subject of this thread.
There actually are manufacturers that publish their measurement protocols. Check Seas and Scan Speak.
roger2,
basically an answer to your question is, it's irrelevant. Manufacturer T/S parameters are merely there to get a ballpark figure what is to be expected of these in order to simshape the FR around its resonant frequency. Once they are broken in, Fs drops a couple of Herz. Either way, it is cosidered the safest practice to measure anyway any unit you intend to work with. It will save you from drawing wrong conclusions.
basically an answer to your question is, it's irrelevant. Manufacturer T/S parameters are merely there to get a ballpark figure what is to be expected of these in order to simshape the FR around its resonant frequency. Once they are broken in, Fs drops a couple of Herz. Either way, it is cosidered the safest practice to measure anyway any unit you intend to work with. It will save you from drawing wrong conclusions.
How accurate manufacturer T/S parameters are depends on the manufacturer.
First there is the issue of quality control. I have measured thousands of drivers. A good make can meet tolerances of less than plus/minus 10%, poorer makes can have dramatically higher variance.
I have found that using the factory numbers as a starting point is much more useful than using those i measure. I have one of the best T/S parameter measuring kits short of the pricey stuff at the factory. It is useful for matching but typically using the factory numbers works better. I am rarely dealing with drivers that don’t come from a good manufacturer.
What Sonce says has some merit, but also significant falsehood. A typical driver will have its suspension loosened sufficiently in a very short time for T/S paarameters but he is VERY wrong about the need for lots of time on some drivers for them to best develop their capabilities to reproduce the small stuff. And if not properly broken in the suspension will get a “memory” which will obscure that small detail forever.
The routine of at least one manufacturer is to choose a random sample of say 50 drivers (out of 1000s), give them a short break-in, measure them (with factory software you are unlikely to ever see) and then publishes the average. I have never had to second guess their numbers.
Also note that unless measured in a pressure/temperature controlled environment the measures for the same driver will be different every day depending on the weather.
dave
First there is the issue of quality control. I have measured thousands of drivers. A good make can meet tolerances of less than plus/minus 10%, poorer makes can have dramatically higher variance.
I have found that using the factory numbers as a starting point is much more useful than using those i measure. I have one of the best T/S parameter measuring kits short of the pricey stuff at the factory. It is useful for matching but typically using the factory numbers works better. I am rarely dealing with drivers that don’t come from a good manufacturer.
What Sonce says has some merit, but also significant falsehood. A typical driver will have its suspension loosened sufficiently in a very short time for T/S paarameters but he is VERY wrong about the need for lots of time on some drivers for them to best develop their capabilities to reproduce the small stuff. And if not properly broken in the suspension will get a “memory” which will obscure that small detail forever.
The routine of at least one manufacturer is to choose a random sample of say 50 drivers (out of 1000s), give them a short break-in, measure them (with factory software you are unlikely to ever see) and then publishes the average. I have never had to second guess their numbers.
Also note that unless measured in a pressure/temperature controlled environment the measures for the same driver will be different every day depending on the weather.
dave
As Dave notes, this varies from manufacturer to manufacturer, although typically the latter is more common.
Two points to keep in mind on this subject however. And I strongly advise you to do so.
Firstly, assuming you have a broken-in drive unit, T/S parameters are not fixed, but vary with drive level, ambient climatic conditions and several other factors. That is one reason Klippel etc. is as comprehensive as it is. T/S parameters per se are technically speaking low-voltage, but that isn't necessarily representative of actual use. Neither are some of the high voltage measures for that matter; a mean usually ends up the most accurate to what is actually happening.
Secondly: having dealt with a few manufacturers for work purposes, I can tell you that many, including some of the big names, base their QA / QC assessment on 10% deviation to Fs. That's it. Nothing else. And if that doesn't alarm you: it should. I've had drivers through here tight to Fs but everything else so far out (not just to the advertised data but in terms of unit-to-unit matching) my patience and sense of humour alike went out of the window. Which, as far as their practical use was concerned, was what I might as well have done with the drivers.
Last edited:
So, "the...driver will be different every day, depending on the weather". So no doubt this is an audible difference, a 0.38 Db hump at 67 hertz because it is a warm & humid day gives you a "punchy, power bass response"??
You give the human ear more credit than its worth.
----------------------------------------------------------------------------Rick.......
You give the human ear more credit than its worth.
----------------------------------------------------------------------------Rick.......
So just whos T/S parameters are the most consistent, & conversely, who tends to have the most inconsistent according to the data?
I don't want to be on the receiving end of bad data, we do not all have the equipment to verify.
----------------------------------------------------------------------------Rick......
I don't want to be on the receiving end of bad data, we do not all have the equipment to verify.
----------------------------------------------------------------------------Rick......
Last edited:
Yes, to a certain extent.
Please indicate where I, Dave, or anybody else said anything of the kind.
Last edited:
Frankly, you should; you can get a good idea with a project box, a couple of resistors, a half-decent soundcard and a copy of REW, AudioTester or similar if you don't want to spend too much.
Generally speaking, to give a few examples, Seas tend to be relatively reliable -more than some anyway. Dayton often are quite good too. Scan are usually consistent in my experience, but not necessarily always quite to advertised. Unit to unit they're usually good though. However, as noted, many 3rd party measures also tend to be done under low voltage drive conditions, so are not automatically more accurate, just one extra data point / snapshot, so something to take into account.
Achieving a relative level of mechanical stability such that any minor variations from that point are caused by changes in the prevailing climatic conditions, VC heating, and, over the long term, gradual component degredation.
On the contrary, I specifically address that 2-hour breaking-in is enough, not the often-quoted "audiophile certified" 100-hour madness:
Also, pink noise signal guarantees there will be no "memory" effect at all.
Thanks for the links
As for SEAS, they do a 2 minute "run-in" before taking readings. Seems not long enough to me. I would think that parameters could change more with further break-in by the consumer. Not knocking them, I know SEAS is a reputable manufacturer of quality drivers.
From the SEAS website:
"Before measuring the T/S-parameters the driver has been run in at near maximum excursion in free air using a sine signal at f0 for 2 min. The resonance frequency and impedance at resonance are measured using a 2 VRMS sine signal.
The moving mass and the BL-product are measured with the laser equipment from Klippel GmbH."
The Scan-Speak burn-in method appears to be similar to SEAS' but without a specific time period specified:
"Before starting measurement of the parameters, the drive unit must be burned-in by exposing it to a (preferably) sine tone at a frequency slightly below or at the initial resonance frequency and then applying a voltage large enough for the driver to move almost to its maximum excursion. If the driver starts clipping, which is clearly audible, turn back the voltage. The T/S parameters should always be measured on a burned-in driver, because the compliance of a completely new driver will change quite a bit in the first few minutes of excitation. So if the data is to be used for system design, the burned-in dataset is the useful one"
So these two well regarded manufacturers of higher-end drivers (SEAS & Scan-Speak) do make the effort to burn-in drivers for the purpose of obtaining T/S values for publishing.
What about other driver manufacturers? For example, Peerless by Tymphany or other mid-level stuff coming out of China?
Thank you for your reply Lojzek
I will agree to disagree that my question is irrelevant. It is relevant to me and my understanding of typical industry methods and procedures, and could likely have an effect on my own measuring procedures in the future. But that is not the topic of this thread.
I will ask you this though ...
Is Fs the only T/S parameter that changes significantly with mechanical break-in? Or do other parameters change as well?
Eighteen Sound, professional drivers from Italy: "Thiele-Small parameters are measured after a high level 20 Hz sine wave preconditioning test."
Tymphany/Peerless: "Measured in Free Air without preconditioning, therefore subject to some deviation." Practically useless.
ScanSpeak: "Before starting measurement of the parameters, the drive unit must be burned-in by exposing it to a (preferably) sine tone at a frequency slightly below or at the initial resonance frequency and then applying a voltage large enough for the driver to move almost to its maximum excursion."
I would say the correct way of burning-in is: 2-hours with pink noise within operating range, close to but less than Xmax.
Tymphany/Peerless: "Measured in Free Air without preconditioning, therefore subject to some deviation." Practically useless.
ScanSpeak: "Before starting measurement of the parameters, the drive unit must be burned-in by exposing it to a (preferably) sine tone at a frequency slightly below or at the initial resonance frequency and then applying a voltage large enough for the driver to move almost to its maximum excursion."
I would say the correct way of burning-in is: 2-hours with pink noise within operating range, close to but less than Xmax.