WTD: Pair of Celestion HF2000 tweeters (must meet strict criteria)

[ToTo Man]

I'm looking for a pair of Celestion HF2000 tweeters that I can store away as backups for my Ditton 66.

However they must meet very strict criteria, because over the last decade or so I have procured some five pairs or so from eBay, and of these 10 units I have only been able to match up 2 that are within ± 1.5dB across their 6kHz-20kHz operating range. Some are out by as much as 5dB or even 10dB in a couple of cases!).

So, if you have a pair of HF2000's and can prove with measurements that they are matched to ±1.5dB from 6kHz-20kHz, please get in touch as I'd be willing to offer a generous sum for them.

Cheers :)

[vintagesteve]

At frequencies above 6k, and certainly over 10k, I would doubt my ability to detect a difference of a 2, 3, 4 or even 5 dB on a stereo recording. In the mid range, yes for sure and also at lower frequencies. But I fully understand what you're trying to achieve. It's a big ask that someone is in possession of that level of data, but best of luck.

[ToTo Man]

At frequencies above 6k, and certainly over 10k, I would doubt my ability to detect a difference of a 2, 3, 4 or even 5 dB on a stereo recording. In the mid range, yes for sure and also at lower frequencies. But I fully understand what you're trying to achieve. It's a big ask that someone is in possession of that level of data, but best of luck.

I know, but I can only live in hope! I'm unfortunately cursed with the ability to hear very small differences between left and right channels, even at high frequencies, which does not fit very well with my love of vintage hifi. Luckily I already have a pair of HF2000's that are pretty well matched, so I don't mind playing the long game. I expect I'll still be 'bumping' this thread a year from now! :D

[Lawrence001]

Very long shot this. You might miss out on a tweeter that while it might not match it's own pair it could match one of yours. If people don't have the equipment you could maybe pay postage both ways to check them yourself?

Sent from my BLN-L21 using Tapatalk

[Mixc]

I'm looking for a pair of Celestion HF2000 tweeters that I can store away as backups for my Ditton 66.

However they must meet very strict criteria, because over the last decade or so I have procured some five pairs or so from eBay, and of these 10 units I have only been able to match up 2 that are within ± 1.5dB across their 6kHz-20kHz operating range (some are out by as much as 5dBs!).

So, if you have a pair of HF2000's and can prove with measurements that they are matched to ±1.5dB from 6kHz-20kHz, please get in touch as I'd be willing to offer a generous sum for them.

Cheers :)

Hi Richard, how are you testing the response?

[ToTo Man]

Hi Richard, how are you testing the response?

Mick, I'm measuring with a UMIK-1 USB mic and REW software using the automatic sine wave sweep with manually selected cut-off points of 2.5kHz and 22kHz. In hindsight I should have used 4kHz as the lower cut-off as there won't be much output below this once connected to the crossover. I imagine Pink Noise would work equivalently well. I kept the levels fairly low (<70dB @ 1m) to prevent risk of damage.

I measured each HF2000 with no crossover, just connected straight to my amp (if you do this you need make sure your amp has no DC offset and you are feeding the tweeter with filtered content because you could damage it with frequencies below its normal operating range).

I placed each tweeter face up on the carpet in the middle of my room, with the mic positioned 1 meter above it on a boom stand. I'm sure those with more knowledge/experience will tell me this isn't the best way to measure, but it's the technique I started off with so I best stick to it or else I'll need to repeat dozens of measurements again!

However you measure, it's vital that the mic isn't moved until all measurements are completed, otherwise all bets are off!

If you're interested you can see my results here:
https://theartofsound.net/forum/showthread.php?59279-Another-Celestion-Ditton-66-Renovation-thread

[Mixc]

Ive just read your post very interesting, So much effort....
To me your measurement procedure should be plenty good enough. Its got my brain ticking, when i run a frequency sweep i can hear the focus of the sound change position (left to right) i put this down to room acoustics or my hearing, maybe my tweeters are not as balanced as they should be.Hmm.
That never ending quest for perfection... Good Luck

[ToTo Man]

Ive just read your post very interesting, So much effort....
To me your measurement procedure should be plenty good enough. Its got my brain ticking, when i run a frequency sweep i can hear the focus of the sound change position (left to right) i put this down to room acoustics or my hearing, maybe my tweeters are not as balanced as they should be.Hmm.
That never ending quest for perfection... Good Luck

Could well be room acoustics. When I measure a complete loudspeaker I measure each one in the exact same location as this takes the room out of the equation. Jump down this rabbit hole at your own peril, as once you have conclusive proof of an imbalance, your subconsciously far more likely to notice it and have the gnawing urge to remedy it!

[vintagesteve]

Just a thought, and I'm not trying to throw a spanner in the works, if you're placing the tweeters down on carpet (i've not seen the carpet, thickness etc) then there's a possibility that subsequent tweeters placed for testing may be tilted slightly in comparison, a carpet not being a stable, repeatable support. This off axis effect would adversely affect the level of signals received by the mic, giving rise to a false comparison. You may have taken this into account of course with how the tweeter is 'placed' on the carpet.

[Lawrence001]

Very good point if you're rounding to 0.5db could make all the difference.

[ToTo Man]

Just a thought, and I'm not trying to throw a spanner in the works, if you're placing the tweeters down on carpet (i've not seen the carpet, thickness etc) then there's a possibility that subsequent tweeters placed for testing may be tilted slightly in comparison, a carpet not being a stable, repeatable support. This off axis effect would adversely affect the level of signals received by the mic, giving rise to a false comparison. You may have taken this into account of course with how the tweeter is 'placed' on the carpet.


Very good point if you're rounding to 0.5db could make all the difference.

That is a good point, but I did try to get each one as levelled as possible to ensure consistency. I suppose a shift of a couple of degrees could affect the response by a few tenths of a dB, but most of the differences between the units I measured were of the order of a few dBs not tenths.

It wasn't so much the variations in overall output level that surprised me (these could easily be balanced with a resistor), but that some of the units had significant dips in their response. Luckily the two units I paired up as the best match for each other also happen to have the most linear response of all the units I tested.

[vintagesteve]

Out of interest, here's the polar response of the Ditton 66 if you haven't already got it. Well, it would be here but it seems there's no facility to upload images any more!

[ToTo Man]

Out of interest, here's the polar response of the Ditton 66 if you haven't already got it. Well, it would be here but it seems there's no facility to upload images any more!

This?

http://i48.tinypic.com/14bthkl.jpg

[vintagesteve]

That's the one, yes.

[ToTo Man]

There doesn't seem to be much affect on the HF response for the first few degrees or so, therefore my measurements should all be comparable.

[ToTo Man]

Bump (still looking for a matched pair of HF2000s).

[ToTo Man]

Another bump to say I'm still looking for HF2000 tweeters. I'm now willing to consider unmatched pairs or single units, in the hope that I might be able to match them up with the units I already have.

[spendorman]

I had a similar problem finding a pair of HF2000s for my LS3/6.

I believe that there are different impedance versions too.

[spendorman]

Interesting

https://www.ebay.co.uk/itm/Pair-of-Celestion-Ditton-Tweeters-Boxed-Unused-551-442-T3365/372608046702?hash=item56c12b966e:g:awwAAOSw~jJcbxQ g:rk:4:pf:0

[ToTo Man]

Interesting

https://www.ebay.co.uk/itm/Pair-of-Celestion-Ditton-Tweeters-Boxed-Unused-551-442-T3365/372608046702?hash=item56c12b966e:g:awwAAOSw~jJcbxQ g:rk:4:pf:0

Nice find! :).

One of the pairs of HF2000 I bought from eBay were NOS and came in the same boxes as shown in the above advert. I couldn't believe my luck, that was until I frequency swept them and discovered they were out by 4dB at certain frequencies :(.

In my experience, from the dozen units I've measured (graphs here (https://theartofsound.net/forum/showthread.php?59279-Another-Celestion-Ditton-66-Renovation-thread)), I struggle to draw a conclusive relationship between their DC resistance and their actual sensitivity. The DCRs of the units I measured range from 3.3 ohms to 4.4 ohms. The units in the above auction have DCRs of around 6 ohms, which would suggest a significantly higher nominal impedance than my existing units. I'm not sure how this would affect the Ditton 66 crossover circuit??

[Patrick Dixon]

When I looked at replacement HF2000s for my 66s a few years ago, I discovered that there are two versions with different DCRs and frequency responses. In the end I replaced them with Hiquophons and rebuilt the crossovers to suit.

No regrets.