http://shop.eadsweden.com/shop/17596/art45/8372645-214090-Jordan_JX60_v1.1.pdf
What I hoped for, like the JXR6 but with a spider. Will have a play on winisd.
What I hoped for, like the JXR6 but with a spider. Will have a play on winisd.
Last edited:
Interesting. Presumably they've updated the range to try to make up some of the ground lost to the Mark Audio drivers. Difficult to make any assessments re the new JX92 models as the plot sizes are different, but it looks like they could be a step forward from their old units. The new JX60 looks a solid little unit too.
One thing that slightly concerns me -apparantly each unit now comes with individual parameters. No bad thing on most levels, but I'm not quite sure what it says about QC tolerances. If they were within a few percent (as they should be, and I trust they will be) that won't be necessary. Still, gives a good impression & it's equally good to see EAD are trying to improve the range rather than simply rely on the name.
One thing that slightly concerns me -apparantly each unit now comes with individual parameters. No bad thing on most levels, but I'm not quite sure what it says about QC tolerances. If they were within a few percent (as they should be, and I trust they will be) that won't be necessary. Still, gives a good impression & it's equally good to see EAD are trying to improve the range rather than simply rely on the name.
Need to see some of those.
Wish the specs were in a more standard format. Based on other specs, efficiency is some almost 85 dB, not bad for a 2". Data for it and the J6T are very close. Wonder how much of the tooling for the J6T that EAD bought off Mark is being used.
Be nice if the data for the new JX92s was in the same format as the original.
The new JX92s have a nice resonance train starting at 5k. 10 dB peak at 8kHz.
dave
Wish the specs were in a more standard format. Based on other specs, efficiency is some almost 85 dB, not bad for a 2". Data for it and the J6T are very close. Wonder how much of the tooling for the J6T that EAD bought off Mark is being used.
Be nice if the data for the new JX92s was in the same format as the original.
The new JX92s have a nice resonance train starting at 5k. 10 dB peak at 8kHz.
dave
Nice? It looks awful. The little Jx60 is ragged in that area too. I have Jx92s's from around 2004. They have the same shape of ripple in the response curve, but not that pronounced. It does not match the published specs for the one they are selling now. I use an active EQ with them for ripple-kill, baffle step, and floor bounce. Gadzooks. I have been tweaking that system for seven years.
Last edited:
I guess the irony in my voice didn't come thru in my post
Another interesting point, it looks like the HD has the specs of what the "regular" one had before EAD bought the company. There was quite a foo-fa-ra when it was 1st suggested that the JX92s specs had become quite different after EAD started building them. Now it has specs very similar to what diyers had measured.
dave
Hi Planet,
I don't think the parameters actually changed, it is just that the way Ted measured them differs from EAD. EAD seems to me a more standard approach. This is the problem with such measurements they have to be seen in context with the method used. If simulator software requires you to type in these parameters they also have to be in-line with the methods the software designer espects.
The new drivers do look a valid improvement but have not heard them yet.
I don't think the parameters actually changed, it is just that the way Ted measured them differs from EAD. EAD seems to me a more standard approach. This is the problem with such measurements they have to be seen in context with the method used. If simulator software requires you to type in these parameters they also have to be in-line with the methods the software designer espects.
The new drivers do look a valid improvement but have not heard them yet.
This problem is not specific to Jordan.
Have a close look,for instance, at the T/S params measurement published at Zaphaudio and among tens of tested models it is quite hard to find any match with official manufacturer's data. Scanspeak o Seas drivers with measured Qts over 0.5 when should be around 0.3 , with accordingly higher Fs, lower Vas, etc...
Have a look also at MJK measurements of the Lowther DX series and compare with the data from the manufacturer...
Part of it is that a lot of the home / small pro measurements tend to be done using a low voltage. In an ideal world, a drive unit's specs. shouldn't be affected by that; many are though. With that said, even when this is accounted for, it's obvious that some manufacturers are being a trifle 'optimistic.' You also have to keep in mind the production QC tolerances & what the various manufacturers work to. Some have a much broader tolerance range than others. Average seems to be about +/- 10% which IMO is already too broad; some clearly are using +/-20%. The good ones are within a ~5% tolerance zone. Alas, there aren't all that many of those.
This situation can be further compounded by the differences you see in measurements, which can often be radically different to published data. There's a few reasons again for that -the manufacturer may have smoothed it off (some is almost always going to be necessary -if most people saw the genuinely raw data, they'd never buy anything), the aforementioned tolerances may come into play, or they may be being optimistic with the truth (again). That doesn't mean that independent measurements are without their problems though. For a start, a really good, lab. quality microphone costs thousands. Literally. When you start getting into really precise measurements, it costs a fortune. The cheaper equipment, even if claimed to be flat from x - y usually isn't to the same level when you look at the details. Nor do most small companies or home-measurers have access to an anechoic chamber, or the precision jigs to set the test gear up optimally. If they lack an anechoic chamber, they also almost certinly lack an IEC standard baffle or baffles too. An interesting example of this cropped up a few days ago when it was noted that a magazine test appeared to be indicating resonances which apparantly shouldn't be there. Dodgy manufacturer data again? Possibly, but in this case, I suspect the issue may lie in the magazine test gear not being perfectly aligned, or the baffle cut-out for the driver not having been chamfered, resulting in reflections back through the cone. That's a guess, true, but there is some indirect evidence to suggest it as at least a possibility.
Six of one, half a dozen of the other. Who do you believe? As a rule, the 3rd party data is extremely useful to have, but I wouldn't take it as perfectly gospel any more than I would most manufacturers. Best policy IMO is to use both, gather as much additional information you can on the driver & the manufacturer generally and draw what conclusions you can with all this in mind. And preferably hear a pair before stumping up your money.
You also have to keep in mind the production QC tolerances & what the various manufacturers work to. Some have a much broader tolerance range than others. Average seems to be about +/- 10% which IMO is already too broad; some clearly are using +/-20%. The good ones are within a ~5% tolerance zone. Alas, there aren't all that many of those. This situation can be further compounded by the differences you see in measurements, which can often be radically different to published data. There's a few reasons again for that -the manufacturer may have smoothed it off (some is almost always going to be necessary -if most people saw the genuinely raw data, they'd never buy anything), the aforementioned tolerances may come into play, or they may be being optimistic with the truth (again). That doesn't mean that independent measurements are without their problems though. For a start, a really good, lab. quality microphone costs thousands. Literally. When you start getting into really precise measurements, it costs a fortune. The cheaper equipment, even if claimed to be flat from x - y usually isn't to the same level when you look at the details. Nor do most small companies or home-measurers have access to an anechoic chamber, or the precision jigs to set the test gear up optimally. If they lack an anechoic chamber, they also almost certinly lack an IEC standard baffle or baffles too. An interesting example of this cropped up a few days ago when it was noted that a magazine test appeared to be indicating resonances which apparantly shouldn't be there. Dodgy manufacturer data again? Possibly, but in this case, I suspect the issue may lie in the magazine test gear not being perfectly aligned, or the baffle cut-out for the driver not having been chamfered, resulting in reflections back through the cone. That's a guess, true, but there is some indirect evidence to suggest it as at least a possibility.
Six of one, half a dozen of the other. Who do you believe? As a rule, the 3rd party data is extremely useful to have, but I wouldn't take it as perfectly gospel any more than I would most manufacturers. Best policy IMO is to use both, gather as much additional information you can on the driver & the manufacturer generally and draw what conclusions you can with all this in mind. And preferably hear a pair before stumping up your money.
This problem is not specific to Jordan.
Have a close look,for instance, at the T/S params measurement published at Zaphaudio and among tens of tested models it is quite hard to find any match with official manufacturer's data. Scanspeak o Seas drivers with measured Qts over 0.5 when should be around 0.3 , with accordingly higher Fs, lower Vas, etc...
The problem isn't a problem (at least not in the sense that you think). T/S are not scalar parameters, but curves, largely a function of drive.
Attached is measure of exactly the same drivers, measured by me (with kit similar to Zaph's) and by Mark Fenlon with industry standard kit. Both sets of data are valid.
If a driver measures similar with each all it means is that the driver has a flat horizontal T/S curve
dave
Attachments
You mentioned that in your email to me Dave, and I didn't quite understand what that means. For instance, Fs should be a hard number. Shouldn't it?
Here's a measurement of the FR88 I showed you:
It's almost perfectly in line with Zaph's measurements with Fs around 120hz. The manufacture says Fs is 95hz I believe. Well, the 3rd parties here (me and zaph) don't have nearly as good equipment as Fountek, but I can understand why a manufacture would want to "claim" a lower Fs than actual. It makes the driver look like a better performer.
At which frequency the speaker resonates should be predictable and is required to properly design boxes, I think. I'm really not good at designing boxes and new to measuring impedance, so maybe you could shed some light on this. Thanks.
Here's a measurement of the FR88 I showed you:
An externally hosted image should be here but it was not working when we last tested it.
It's almost perfectly in line with Zaph's measurements with Fs around 120hz. The manufacture says Fs is 95hz I believe. Well, the 3rd parties here (me and zaph) don't have nearly as good equipment as Fountek, but I can understand why a manufacture would want to "claim" a lower Fs than actual. It makes the driver look like a better performer.
At which frequency the speaker resonates should be predictable and is required to properly design boxes, I think. I'm really not good at designing boxes and new to measuring impedance, so maybe you could shed some light on this. Thanks.
Yes i know that... the parameter game
:http://www.ejjordan.co.uk/PDFs/parameter_game.pdf
The problem is that most diyers have been "educated" to feed their simulators with those "scalar" values without questionning and feel quite unconfortable ... and confused ... when a sophisticated gentleman tries to explain that things made so easy by popular programs such as WinIsd ( or Woofer Tester...) are not so easy... or simply " drive level dependant"... Do you mean that the liters i need for my cabs are also "drive level dependant"?
Many dyers just jump from pure simulation straight to the "real wood veneered" final execution without any curiosity for experimenting with cheap mdf different prototypes to test something real. Why loosing time testing a proto when a simple simulation will tell you the whole story ?
Last edited:
That is a shocker for a manufacturer to be so far out with the truth. Same with Tangband response graphs.
btw I really think Zaph's site is the dog's..its good to read so many impartial tests -big fan of Zaph, pls review the new Jordans for us
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.