this ebay listing for 4 faulty tweeters shows the 2 different types you described that I am used to seeing:
Celestion HF 2000 TWEETERS DITTON 66/44/25 X4 faulty on eBay (end time 06-Jul-10 19:56:00 BST)
Hi,
I found my way here as I'm investigating HF2000 replacement for the Ditton 44. One of mine got toasted by a cartridge preamp which decided to turn into an oscillator.
Now I've taken a few measurements, I find the "working" one has a DC resistance of 9 ohms and the impedance at 5 kHz is 10.3 ohms.
Combine that and sba's measurements of four of them and I'm left thinking that buying s/h HF2000s is risky and could get expensive!
I've autopsied my dead HF2000. It's fairly clear that tweeter design has advanced significantly since the days of the HF2000. Computer modelling, improved measurement techniques and more consistent manufacturing techniques results in better (and less fragile) drivers.
(Now I've seen for myself just how fragile the HF2000 is, I'd be worried about frying exact replacements. I enjoy music; worrying about my tweeters is a distraction I'd rather not have!)
I've looked at the data for a few tweeters and agree that the SEAS 19TFF1 ought to be suitable (with crossover changes to integrate it with the system properly.) I've ordered a pair, hopefully they'll turn up next week.
I've also traced out the 44 crossover and measured the inductors. The tweeter section is identical to that shown on Celestion's 66 diagram. (If anyone's interested I'll post that.) So I think that any tweeter replacement and crossover changes which work for the 66 would apply equally well to the 44.
Anyway, that's why I'm here. I'll close with some photos from the autopsy.
Regards, Kat
I found my way here as I'm investigating HF2000 replacement for the Ditton 44. One of mine got toasted by a cartridge preamp which decided to turn into an oscillator.
Now I've taken a few measurements, I find the "working" one has a DC resistance of 9 ohms and the impedance at 5 kHz is 10.3 ohms.
Combine that and sba's measurements of four of them and I'm left thinking that buying s/h HF2000s is risky and could get expensive!
I've autopsied my dead HF2000. It's fairly clear that tweeter design has advanced significantly since the days of the HF2000. Computer modelling, improved measurement techniques and more consistent manufacturing techniques results in better (and less fragile) drivers.
(Now I've seen for myself just how fragile the HF2000 is, I'd be worried about frying exact replacements. I enjoy music; worrying about my tweeters is a distraction I'd rather not have!)
I've looked at the data for a few tweeters and agree that the SEAS 19TFF1 ought to be suitable (with crossover changes to integrate it with the system properly.) I've ordered a pair, hopefully they'll turn up next week.
I've also traced out the 44 crossover and measured the inductors. The tweeter section is identical to that shown on Celestion's 66 diagram. (If anyone's interested I'll post that.) So I think that any tweeter replacement and crossover changes which work for the 66 would apply equally well to the 44.
Anyway, that's why I'm here. I'll close with some photos from the autopsy.
Regards, Kat
Attachments
66 woofer
Here are some of my first measurements. The software program would only display three measurements at a time, so the polars are spread out among several graphs. The measurement distance was four feet. It appears that the ditton 25 is really pushing it by using this woofer up to 2khz.
Here are some of my first measurements. The software program would only display three measurements at a time, so the polars are spread out among several graphs. The measurement distance was four feet. It appears that the ditton 25 is really pushing it by using this woofer up to 2khz.
Tweeters
Ignore post #486 about the JBL tweeter...that tweeter had a damaged cone. The orange plot here shows an undamaged JBL cone tweeter. The blue plot shows the HF2000 dome tweeter. The purple plot shows another Hf2000, the earlier version with the small diameter cardboard mounting. The impedance plot is for the blue colored HF2000 (I forgot to save the imedance plot for the other hf2000). All measurements at 4', with 1/8 octave smoothing
Ignore post #486 about the JBL tweeter...that tweeter had a damaged cone. The orange plot here shows an undamaged JBL cone tweeter. The blue plot shows the HF2000 dome tweeter. The purple plot shows another Hf2000, the earlier version with the small diameter cardboard mounting. The impedance plot is for the blue colored HF2000 (I forgot to save the imedance plot for the other hf2000). All measurements at 4', with 1/8 octave smoothing
Last edited:
that 1/8 th Octave smoothing ... and other matters
Hi sba,
some good stuff in your plots !
There is a LOT I can comment on, and I will return and do so as soon as I have time available, but for now:-
the 1/8th octave smoothing has messed things up I think ...
better would be to see plots of all the mid-domes and tweeters with no smoothing ... if you still have such on file ... ?
That Impedance Plot for the HF2000 tweeter does not match any of your 2009 tweeter Impedance plots !
Is this new Plot a measurement in a different circuit connection ?
The Tabulated Data is missing for two of the JBL tweeters
- for the two which have Z(max) at about 10.9 <---> 11 ohms.
Do you have a Relative Sensitivity {in dB} comparison between the Celestion and the JBL tweeters ?
Are they all Le25, or are some Le26 ?
... as you mentioned Le26 in #478 on Page 48.
There is scope for easy crossover mods to match some of the JBL tweeters to the Celestion mids
- I'll post more about this next time, but knowing the dB difference then will assist.
Hi sba,
some good stuff in your plots !
There is a LOT I can comment on, and I will return and do so as soon as I have time available, but for now:-
the 1/8th octave smoothing has messed things up I think ...
better would be to see plots of all the mid-domes and tweeters with no smoothing ... if you still have such on file ... ?
That Impedance Plot for the HF2000 tweeter does not match any of your 2009 tweeter Impedance plots !
Is this new Plot a measurement in a different circuit connection ?
The Tabulated Data is missing for two of the JBL tweeters
- for the two which have Z(max) at about 10.9 <---> 11 ohms.
Do you have a Relative Sensitivity {in dB} comparison between the Celestion and the JBL tweeters ?
Are they all Le25, or are some Le26 ?
... as you mentioned Le26 in #478 on Page 48.
There is scope for easy crossover mods to match some of the JBL tweeters to the Celestion mids
- I'll post more about this next time, but knowing the dB difference then will assist.
mods for Celestion 44 ... and about Celestion tweeters
Hi kayelem,
I am currently discussing adapting SEAS 19TFF1 to Celestion 44 with another owner, thus there may be more I can say about that later.
For now, if you have not already found it, then Search in this Forum for the Thread titled:-
Crossover nightmare!!!!!!!
started by:-
lorienblack
Not all of it will be of use to you, but some of it may be.
You can Post in that thread about crossover mods for the 44 if you want to
- I will reply there.
If you would like to Post the exact mH measured values of your inductors - in either Thread - that could be of some use because Tolerance variations do exist and I'm interested to be able to allow for whatever amount there is.
****************
Thankyou for posting the photos of the HF2000 types.
Some recent information I found leads me to think there may not have been a HF1000 back at the time of HF2000.
After HF2000, and partly concurrent with it, was the HD1000
{which readers can find in that Celestion Archive link in #472 on Page 48, courtesy of ittf !}.
There was a Celestion HF1000 introduced a few years later.
Rogers used it in their LS7 loudspeakers initally
- before the application of the metal dome tweeter there to LS7t specification.
Your 9 ohm HF2000 ... way back before half-way through this Thread is a Post by a member who has one 9 ohm HF2000.
Now as two such mutations in existance perhaps one of you may be able to use both !
A simple mod to the crossover will allow the 9 ohm version to be adapted.
I'll post about that if anyone is interested.
Hi kayelem,
I am currently discussing adapting SEAS 19TFF1 to Celestion 44 with another owner, thus there may be more I can say about that later.
For now, if you have not already found it, then Search in this Forum for the Thread titled:-
Crossover nightmare!!!!!!!
started by:-
lorienblack
Not all of it will be of use to you, but some of it may be.
You can Post in that thread about crossover mods for the 44 if you want to
- I will reply there.
If you would like to Post the exact mH measured values of your inductors - in either Thread - that could be of some use because Tolerance variations do exist and I'm interested to be able to allow for whatever amount there is.
****************
Thankyou for posting the photos of the HF2000 types.
Some recent information I found leads me to think there may not have been a HF1000 back at the time of HF2000.
After HF2000, and partly concurrent with it, was the HD1000
{which readers can find in that Celestion Archive link in #472 on Page 48, courtesy of ittf !}.
There was a Celestion HF1000 introduced a few years later.
Rogers used it in their LS7 loudspeakers initally
- before the application of the metal dome tweeter there to LS7t specification.
Your 9 ohm HF2000 ... way back before half-way through this Thread is a Post by a member who has one 9 ohm HF2000.
Now as two such mutations in existance perhaps one of you may be able to use both !
A simple mod to the crossover will allow the 9 ohm version to be adapted.
I'll post about that if anyone is interested.
Last edited:
I can rerun the data. Here's what some of the options look like. The top is 1/24, the middle is 1/8, and the bottom is "impulse time window". Let me know your preference!
Yes, it was made last week using a different computer. The mid dome impedance plot is also fresh.
I'm out of time, and will return with this info.
Well, I just learned how to adjust the impulse window to see reflections. In this plot of a single 66 woofer, there’s a big one around 2ms, which translates to about 27 inches. [ Frequency, Wavelength, Time and Distance Calculator ]
And that’s not supposed to be there (I think)… I need to re-examine my setup for problems.
These are different curves of the same woofer--
The blue curve utilizes the FR “impulse-time window” with the “auto gate” function.
The red curve is the same window, but with the gating manually adjusted to 400hz. It appears that including those 2ms reflections makes the upper part of the curve less useful.
The green curve is the FR with 1/8 octave amplitude smoothing.
***********************
I also discovered an error on all of my off-axis measurements. The 0 and 60 degree ones are correct, but the ones marked 30 degrees are actually 37.5, and the ones marked 45 are actually 52.5
I think I have to sort out the reflection issues, and then retest the drivers….It was unlikely that I’d get all this stuff right on my first attempt.
And that’s not supposed to be there (I think)… I need to re-examine my setup for problems.
These are different curves of the same woofer--
The blue curve utilizes the FR “impulse-time window” with the “auto gate” function.
The red curve is the same window, but with the gating manually adjusted to 400hz. It appears that including those 2ms reflections makes the upper part of the curve less useful.
The green curve is the FR with 1/8 octave amplitude smoothing.
***********************
I also discovered an error on all of my off-axis measurements. The 0 and 60 degree ones are correct, but the ones marked 30 degrees are actually 37.5, and the ones marked 45 are actually 52.5
I think I have to sort out the reflection issues, and then retest the drivers….It was unlikely that I’d get all this stuff right on my first attempt.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.