For those that always questioned why Lambda and AE didn't put a bunch of effort into detailed testing for the public now you can see some of the issues. I doubt that 2 speaker companies do their testing the exact same way, no matter what there is a compromise. I even went as far as setting up in the middle of a soccer field once only to pick up LF from the road and wind problems.
This concrete issue is a new one on me
BTW if Earl does help with this I would be most grateful. I am always willing to learn what is found and if any problems can they be fixed. The Johnny Holmes phase plugs we had to quickly make for the Apollo 15" drivers comes to mind, who would have thought people would be using 15" drivers for midranges back then.
This concrete issue is a new one on me
BTW if Earl does help with this I would be most grateful. I am always willing to learn what is found and if any problems can they be fixed. The Johnny Holmes phase plugs we had to quickly make for the Apollo 15" drivers comes to mind, who would have thought people would be using 15" drivers for midranges back then.
soongsc said:
A large baffle is needed, because this 15 incher can go lower in frequency than most 7 inchers and under that Zaph usually test, so you want to avoid baffle step problems...
Since most near field measurements are made against the dustcap, that would put the 15 incher at a lower sensitivity versus competition since the phase plug is in the way.
Vas added mass method sucks, Zaph agrees on this and specifically explain it on his website to never do that.
Zaph baffle is quite stiff and heavy, but yeah, it will vibrate anyway I think... you would need to test at a really low level... but that would be a compromise with the noise floor that is achieved...
gedlee said:
Hi Earl,
For ground plane measurements, obviously the size of the microphone diaphram/inlet is an issue above some high frequency, and the enclosure or mounting surface makes for some aberration which is more significant at close distances relative to the size/frequency.
For the case of a non-deal boundary rigidity, how lossy would the ground or other surface have to be to result in a 0.5, 1 or 2dB deviation?
Mark Seaton said:
That would have to be tested right? What concerns me about ground plane is the difusion of the wavefront as it propagates over a rough surface. What's "rough"? depends on the wavelength and the surface. My point is that I have seen ground plane measurements on speakers that I have also seen as free field and they are not the same. Free field is hard to argue with as virtually no assumptions are made, but ground plane has a BIG assumption about the reflectivity of the plane. In our chamber back at Ford we saw aberations of the results at the 90 degree position (speaker in the floor, pointing up) and the floor was solid and fairly smooth. The travel distance was only 1 meter, and yet this was enough to affect the measurement.
To me its up to the ground plane guys to prove that this method works. That it gives the same answers as a free field, which has to be assumed to be the standard. This has never been done to my knowledge and so I have to question the data until such time as its shown to be valid. If you have such data, or anyone else does either, then I'd love to see it.
simon5 said:
Either method has its errors. Most errors come from the perturbation not being what you think that it is. I showed in a patent how one could get the Thiele-Small data without a perturbation with a single data excitation, but three data channels, voltage, current and pressure. From these three you can get whatever data you need without a perturbation of the system. It can also be done with two runs, one of voltage to pressure and then current to pressure.
I think that there is no perfect way of measuring driver performance. The method of extracting the parameters actually plays a very important role. Also the person making the measurements should be aware of what to expect when a specific measurement method is used. I have found that using delta mass method generated Vas is quite accurate for design resulting in Qb and Fb in vented enclosures within 5% toloerance measuring the finished product. So I'm not sure what kind of errors you finally ended up with in your design experience because the graphs in the above link does clearly show the calculated T/S.John_E_Janowitz said:
The response of all drivers will measure differently when you go from near field to far field. In the drivers that I have measured, near field actually reduces edge diffraction effects resulting in measurements more closely ressembling the piston mode with the cone breakup modes. This is reasonable due to the distance ratio between the measured point and the edge to the mic respectively. However, if the mic is not near enough, of course there will be diffraction problems.
I have not reviewed EV data, nor do I know the details of this sealed enclosure you mention; but if it's not build like an anechoic chamber, then it's problematic.
This kind of ground plane measurement has it's limitation. I recall the stereophile or some other on-line magazine had a series of articles on it you probably want to review before you spend too much effort in this kind of setup. There is no perfect measurement method for all, just some methods best for particular situations.
So you are saying that data can be provided in private? Klippel data?nickmckinney said:For those that always questioned why Lambda and AE didn't put a bunch of effort into detailed testing for the public now you can see some of the issues. I doubt that 2 speaker companies do their testing the exact same way, no matter what there is a compromise. I even went as far as setting up in the middle of a soccer field once only to pick up LF from the road and wind problems.
This concrete issue is a new one on me
BTW if Earl does help with this I would be most grateful. I am always willing to learn what is found and if any problems can they be fixed. The Johnny Holmes phase plugs we had to quickly make for the Apollo 15" drivers comes to mind, who would have thought people would be using 15" drivers for midranges back then.
soongsc said:
As stated above in the linked thread, you can end up on the right Fb and Qb, but by having a flawed Vas and compensating with Mms, increasing the driver diameter, playing with BL and else so everything falls correctly in place.
Just like when you only have Qts, then you either fix Qes to have Qms or otherwise. It will simulate the same Fb and Qb, but...
Did you try the delta compliance method to check for a variation with your delta mass ? Did you check the Mms on a scale to see if it's right ?
Just wondering, because delta-mass is really frowned upon, and you seem to succeed with it anyway.
I actually did no compensation to obtain the measured Qb and Fb values. The problems mentioned in the link really could be due to many things. If you don't reasonably balance the mass, the results are bad; if you place the mass too far away from the vc location, the results will also be bad. Yes, you have to measure the actual added weight. I use a scale with a resolution of at least 0.1g and 0.01g for very light coned drivers. But since I do not know the conditions of what others use, really hard to tell what ptoblem it causes.simon5 said:
The problem using delta V on a driver with a phase plug is that you now always have leakage around the phase plug.
I've done a lot of work on TS measurements and both methods are suspect.
But lets get back to the discussion topic. I never heard back from John about what I proposed and if this suites him to the point where he would build the box and send it to me along with the speaker so that I could test - and post - the data for his and some B&C drivers that I have.
But lets get back to the discussion topic. I never heard back from John about what I proposed and if this suites him to the point where he would build the box and send it to me along with the speaker so that I could test - and post - the data for his and some B&C drivers that I have.
I think we've noticed here how different people want different taken different ways. To measure everything in every way that people want and measure other drivers in all the same way for comparison sake would take forever.
Earl, I would still like to get you a TD12M or TD15M, whichever you choose. I have had others asking about the TDM's in your kits for sometime now and I believe they would provide added benefit to your kits. Possibly an upgrade at the additional cost difference between what you pay for the B&C. Of course the benefits need to be verified and it needs to be worth it to put the extra time into designing another crossover. Regardless I think the first step for that and for others to see a comparison is to get you a driver.
John
Earl, I would still like to get you a TD12M or TD15M, whichever you choose. I have had others asking about the TDM's in your kits for sometime now and I believe they would provide added benefit to your kits. Possibly an upgrade at the additional cost difference between what you pay for the B&C. Of course the benefits need to be verified and it needs to be worth it to put the extra time into designing another crossover. Regardless I think the first step for that and for others to see a comparison is to get you a driver.
John
John
Sure, I will be doing the most comprehensive data set on the Abbeys that I have done to date in the next few weeks. I will test your 12 and the B&C that I use and post a comparison as I see it. As long as I can get that driver in a week or ten days.
Once I have the data, changing the crossover is no problem and if people want to pay the price difference between the drivers when they order the kits I have no problem at all shiping them the AE driver. They do need to understand that this is going to be a couple hundred dollars per speaker - but its their money.
Sure, I will be doing the most comprehensive data set on the Abbeys that I have done to date in the next few weeks. I will test your 12 and the B&C that I use and post a comparison as I see it. As long as I can get that driver in a week or ten days.
Once I have the data, changing the crossover is no problem and if people want to pay the price difference between the drivers when they order the kits I have no problem at all shiping them the AE driver. They do need to understand that this is going to be a couple hundred dollars per speaker - but its their money.
Really wanted to see the TD15-M with Appollo upgrade since it's the flagship with the best high frequency extension with the curvilinear cone and the highest efficiency, but well...
As tinitus said, it will be very interesting none the less !
I want to congratulate John and gedlee to have found common ground on this and I really hope it will be a success. Can't wait !
As tinitus said, it will be very interesting none the less !
I want to congratulate John and gedlee to have found common ground on this and I really hope it will be a success. Can't wait !
Over all these years, I've bought each and every audio stuff (including drivers of course) by studying the published information only. And I never regret doing so.
Maybe I've been lucky, or it's simply because my personal purchasing is just too few to be meaningful in statistics. But I've been very cautious in spending my (hard earned) money.
All those mentioned above were long before I joint this forum or beyond the discussions available here (it's my own personal little misfortune, it's too late I found here).
And now I've read so much valuable information in this forum. It's far far better than what I had done for my earlier purchasings by just studying the publised stuff. So many "cross-references" are written around here already, aren't they? I mean, there're already many different opinions on various drivers. They might be subjective individually, but you can always combine them into your own conclusions. By doing so, it's objetive enough.
So, I will never worry about any testing by whomever, whatever. AE drivers are top quality stuff, no doubt. Once I have enough savings, I WILL buy some AE drivers. That's for sure.
OK, or maybe 18sound ...
Maybe I've been lucky, or it's simply because my personal purchasing is just too few to be meaningful in statistics. But I've been very cautious in spending my (hard earned) money.
All those mentioned above were long before I joint this forum or beyond the discussions available here (it's my own personal little misfortune, it's too late I found here).
And now I've read so much valuable information in this forum. It's far far better than what I had done for my earlier purchasings by just studying the publised stuff. So many "cross-references" are written around here already, aren't they? I mean, there're already many different opinions on various drivers. They might be subjective individually, but you can always combine them into your own conclusions. By doing so, it's objetive enough.
So, I will never worry about any testing by whomever, whatever. AE drivers are top quality stuff, no doubt. Once I have enough savings, I WILL buy some AE drivers. That's for sure.
OK, or maybe 18sound ...
I've got a pair of TD15M-Apollos ordered. I'll be measuring them at some point for crossover etc - if people can clearly and simply state what and how they want measured I'll give it a shot. It won't be until September at the earliest when I get time.
I have an ECM8000 mic, and RME soundcard, and I can use any free software - I have HOLMimpulse.
I have an ECM8000 mic, and RME soundcard, and I can use any free software - I have HOLMimpulse.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.