An Epic Tale Of Two Celestion Ditton 66 Refurbishments

[ToTo Man]

An Epic Tale Of Two Celestion Ditton 66 Refurbishments

My latest Celestion Ditton 66 refurbishment is now complete, meaning I now own two stellar pairs of these majestical loudspeakers. I have chosen to document the refurbishments in this new and hopefully more focused thread than my previous one, in case Admins wish to make it a ‘Sticky’ as a resource to assist others.

What follows is a summary of the issues these speakers had prior to the refurbishment and what the refurbishment entailed.

2018 Ditton 66 pre-refurbishment nearfield:
https://farm2.staticflickr.com/1964/45569433221_b98b9685a8_o.jpg

Issues identified through listening and confirmed by measurements:
- Distorted midrange output and treble in one loudspeaker (measurements confirmed elevated THD levels and also a large resonance at 5kHz).
- Different midrange timbres between left and right speakers (measurements confirmed different MF sensitivities and responses).
- Significant peak at 5kHz in one loudspeaker.
- Different treble timbres between left and right speakers (measurements confirmed different HF responses).
- Louder bass in one loudspeaker (measurements confirmed different LF sensitivities).

2018 Ditton 66 refurbishment process:

1) Drivers:
- 1a) Remove all drive units and measure in free space with sine sweeps to check frequency response and distortion.
- 1b) Source replacement T2619 bass units, MD500 mids and HF2000 tweeters, measure and pair-match to be as close +/- dB tolerance as possible.

The original drive units measured as follows:
https://farm2.staticflickr.com/1911/45569438061_dc8be9fb30_o.jpg
https://farm2.staticflickr.com/1949/45569438191_5658198941_o.jpg
https://farm2.staticflickr.com/1910/45569438261_264e3c2e39_o.jpg

The new drive units I selected to replace the original units measure as follows:
https://farm2.staticflickr.com/1938/45569437701_4e008e2bd4_o.jpg
https://farm2.staticflickr.com/1904/45569437851_f1d4b0314b_o.jpg
https://farm2.staticflickr.com/1924/45569437931_45bdd8f1ce_o.jpg

2) Crossovers:
- 2a) Replace original jump wires on TBC crossovers with solid core 1mm PCOCC.
- 2b) Replace original hookup wires with Van Damme Studio Blue 1.5mm.
- 2c) Replace original capacitors with like-for-like equivalents selected to preserve the original voicing. Measure capacitance on DVM and pair-match caps between crossovers. Measurements revealed all electrolytics to be around 4% higher than printed value, polys were +/- 1%. Printed cap values used as follows: 72uf Alcaps in LF and 4uf Alcap in MF shunt positions, 22uf Mundorf ECap + 1uf Ansar Supersound in MF circuit, 4uf and 6.2uf Ansar Supersound in HF circuit.

3) Binding posts:
- Replace original binding posts with 4mm 5-way gold-plated solid brass posts.

4) Damping:
- Replace original but mismatched open-cell foam damping with other original but matched open-cell foam damping. (I discovered one cabinet contained 40mm thick foam but the other cabinet contained a mixture of 40mm and 52mm. I’m not sure why, but perhaps thicker damping was used to compensate for one bass driver being louder than the other? I replaced everything with 40mm thickness, pilfered from another pair of Ditton 66 enclosures!

5) Test and listen:
- I allowed the speakers to burn-in with pink noise for a minimum of 12 hours before I measured or performed any critical listening.

- Having owned Ditton 66 for more than a decade, I know that they are quite fussy about listening height, especially in modest-sized rooms. Unless you slouch on a beanbag or sit a considerable distance from the speakers, you simply can’t just plop them on the floor and expect a smooth response through the mids and highs. In my experience, the HF2000 should ideally be raised above ear level, otherwise you will experience a combination of comb filtering between the MD500 and HF2000 units and diffraction off the top lip of the cabinet, causing a recessed response in the upper mid and lower treble frequencies that becomes progressively worse the lower the drive units are relative to your ears and the closer your listening position is to the speakers. You can clearly see the effect in the nearfield measurements I previously took of my 2015 Ditton 66:

2015 Ditton 66 post-refurbishment nearfield at varied heights:
https://farm2.staticflickr.com/1826/43444688612_4336bb918a_b.jpg

- I therefore measured each 2018 Ditton 66 loudspeaker on the 185mm high plywood plinth that I normally use when listening to the speakers, and I positioned the microphone at ear height (1.04m). This places the mic almost exactly halfway between the MD500 and HF2000 axes. I also included nearfield measurements on-axis with both the MD500 and HF2000 to re-check the effect of varying the listening height. These measurements once again confirm that it is preferable to align your ears closer to the MD500 unit than the HF2000 if you seek a more linear response:

2018 Ditton 66 post-refurbishment nearfield at varied heights:
https://farm2.staticflickr.com/1939/31708075658_5777b8d7d5_o.jpg

- All measurements were made with the grilles OFF. Like most vintage loudspeakers, the Ditton 66 I presume was designed to be used with the grille on. However, unlike typical wood-frame cloth grilles that usually have a very even and subtle dampening effect on the high frequencies, the Ditton 66 grille is a heavy, fabric-clad metal grid with a rectangular cutout for the HF2000. It effects a significant change in the output above 3kHz, and the change is not entirely even across the frequencies. I was particularly concerned about the grille causing additional diffraction for the nearfield measurements, another reason why I removed it.

- The Ditton 66 was of course not designed to be a nearfield monitor, and the comb filtering between the MD500 and HF2000 becomes less of a concern as the listening distance increases. I have therefore also included farfield measurements taken at my listening position, which is just over 2 metres distance from the speakers. I toed-in the speakers so that their beam hits just behind the listening position (my listening room has been acoustically treated with absorption and I find the imaging and definition on most of my speakers is better with toe-in).

- The farfield measurements, in my opinion, provide a more realistic representation of the Ditton 66 mid and high frequency balance. Once again I have included measurements taken on-axis with both the MD500 and HF2000 to allow comparison of the effect of varying the listening height at the 1 metre nearfield distance vs the 2 metre farfield distance. The seemingly large differences you see between the left and right speakers in the low and mid frequencies in the farfield measurements should be ignored, as my room's layout/construction is asymmetric and therefore the left and right speakers are subject to a different combination of peaks and nulls.

2018 Ditton 66 post-refurbishment nearfield pair-matching:
https://farm2.staticflickr.com/1949/31709466928_aa1f11da98_o.jpg

- As you can see from the nearfield measurements, pair-matching is generally excellent and there are only a few occasions where it diverges by more than 1 dB. My ears are particularly picky when it comes to channel imbalances and I cannot detect any appreciable difference between the two loudspeakers even when isolated and fed a mono signal, and that is a first! I’m pretty sure this is as good as I’ll ever get and I suspect I am probably in possession of one of the best matched pairs of Ditton 66 on the planet!

- I also repeated the measurements with the Ditton 66 I refurbished in 2015 to see how my efforts compare. (My 2015 Ditton 66 use the same model of HF2000 but earlier MF500 and T1600 drivers, and I used the exact same capacitor compliment as in my 2018 Ditton 66 with the exception of the MF circuit, where I fitted 30uf instead of 24uf):

2015 Ditton 66 post-refurbishment nearfield pair-matching:
https://farm2.staticflickr.com/1941/45531310462_0ce822ab58_o.jpg

- When selecting the drive units to use in the 2015 refurbishment, I did not perform any measurements, but instead pair-matched by ear using pink noise. As you can see, I did very well with the bass and mid units but not so well with the tweeters which diverge by as much as 3.5 dBs in places above 6kHz. As I explained in an earlier thread however, I have since discovered much larger variations than this in other HF2000 units, so the pair-matching isn’t nearly as bad as it might have been. While it isn’t noticeable during normal playback, I may re-visit this at some point in the future and see if I can locate a more closely-matched pair of HF2000’s for my 2015 Ditton 66 as I do enjoy a challenge!

- I thought it would be interesting to compare and contrast the tunings of my 2015 and 2018 refurbished Ditton 66 so have overlaid the measured nearfield responses of both in the following graph:


2018 Ditton 66 post-refurbishment nearfield vs 2015 Ditton 66 post-refurbishment nearfield:
https://farm2.staticflickr.com/1931/31709697618_6c63da5261_o.jpg

- The nearfield frequency response of the 2018 Ditton 66 is generally smoother and more neutral through the MF and HF than the 2015 Ditton 66. The 2015 Ditton 66 has a +2dB hump between 550Hz-1kHz and is +4dB hotter above 8kHz. The 2015 Ditton 66 is however more neutral between 300Hz-600Hz. Compared to the published anechoic measurements, I’d say my 2018 Ditton 66 is a closer match to stock specification:

Ditton 66 published anechoic measurements:
https://farm2.staticflickr.com/1917/30707000467_62954f446b_o.jpg

- As I did not take free space measurements of any of the drive units I installed in my 2015 Ditton 66, it is difficult to say whether the frequency response differences that are evident below 5kHz are due to: a) intentional design revisions to the drive units and/or crossovers, b) unavoidable intersample variations, or c) a combination of both. I think c) is the most likely explanation.

- As the HF circuits are identical in the crossovers of my 2015 and 2018 Ditton 66, I am attributing the differences in HF output above 5kHz entirely to intersample variations in the HF2000 units used. During the 2018 refurbishment I measured a total of thirteen HF2000 units to select the closest matching pair, and I found significant differences in both sensitivity and linearity by as much as 8dB is some cases (see https://farm2.staticflickr.com/1803/43444687312_1b1c515378_b.jpg). Finding a closely matched pair of HF2000’s was NOT easy!

2018 Ditton 66 post-refurbishment farfield from listening position (mic at MD500 height):
https://farm2.staticflickr.com/1945/31742778408_4770cac4ab_o.jpg

2018 Ditton 66 post-refurbishment farfield from listening position (mic at HF2000 height):
https://farm2.staticflickr.com/1907/31742778548_dde3b7b1b7_o.jpg

2015 Ditton 66 post-refurbishment farfield from listening position (mic at MF500 height):
https://farm2.staticflickr.com/1918/31742778598_f5d7d5f155_o.jpg

2015 Ditton 66 post-refurbishment farfield from listening position (mic at HF2000 height):
https://farm2.staticflickr.com/1914/31742778478_94ae09d74a_o.jpg

- The farfield measurements show that, as expected, the balance between the mid and high frequencies is less affected by listening height than it is for nearfield. However, to my ears, the tonal balance still sounds better when the MF unit is closer to ear level than the HF unit. The measurements reveal an undesirable null in the response at 700Hz when the mic is at the exact height of the MF unit. This is either being caused by floor-bounce or possibly cancellation between the MF and LF drivers, as the null is significantly diminished when the listening height is varied by a small amount, and it is therefore not a cause for concern IMO.

2018 Ditton 66 post-refurbishment farfield vs 2015 Ditton post-refurbishment farfield from listening position (mic at MD500/MF500 height):
https://farm2.staticflickr.com/1949/31742778678_19d5b223fe_o.jpg

2018 Ditton 66 post-refurbishment farfield vs 2015 Ditton post-refurbishment farfield from listening position (mic at HF2000 height):
https://farm2.staticflickr.com/1959/44701299425_3f9db2b959_o.jpg

- Ignoring the measurements for a moment, and commenting instead on my subjective listening with both speakers, I feel that both the 2018 and 2015 Ditton 66 benefit from some EQ to reduce the output of the lower mid frequencies. I appreciate that it is part of the tuning or ‘character’ of the 66, however I feel the elevated output here adds a wooly warmth that is detrimental to the overall transparency of the speaker. I find the areas requiring EQ to differ slightly between the two models, with the 2015 Ditton 66 benefitting from having the 500Hz-1.25kHz region reduced, and the 2018 Ditton 66 benefiting from extending the area of adjustment down by a couple of hundred Hz to cover 300Hz-1.25kHz.

Conclusions:
- I am delighted with how my 2018 Ditton 66 refurbishment has turned out. I best not contemplate the man-hours and expense that has gone into them, but the proof is in the measurements and listening, and it is clear to me that going the extra mile with pair-matching both the drive units and capacitors has been worth it. I don’t think I’ve heard a closer matching pair of vintage loudspeakers, and before embarking on this project I wouldn’t have thought such close convergence on 40+ year old transducers was possible.

- As for the sound, it really is quite stunning. The 2018 Ditton 66 isn’t as immediately impressive upon first audition as my 2015 Ditton 66 or other references for that matter. As a Tannoy Monitor Gold and Sennheiser HD800S user for many years, I am used to a hotter than neutral top end, so my 2018 Ditton 66 did require a brief period of acclimatisation. Once my ears adjusted to its smoother top end, it was evident that all the detail was still there in spades, it just wasn’t being etched into the soundscape quite as explicitly. I could quite happy listen to this speaker for hours without the worry of listener fatigue, and also without feeling that I’m missing out on detail. It’s not often that a speaker is able to walk this proverbial tight rope without inevitably leaning to one side.

- To sum up my 2018 Ditton 66 in one sentence, I’ll simply say, “you’ll have to prise these mothers from my cold, dead hands!!!…”


.

[ToTo Man]

2018 Ditton 66 Refurbishment Photo Diary:

https://farm2.staticflickr.com/1966/31695397598_1113b99534_o.jpg

https://farm2.staticflickr.com/1943/31695397198_0f78d85e44_o.jpg

https://farm2.staticflickr.com/1977/31695396738_8f284c86a5_o.jpg

https://farm2.staticflickr.com/1980/31695396338_31018ba57f_o.jpg

https://farm2.staticflickr.com/1961/31695395838_b052d74b90_o.jpg

https://farm2.staticflickr.com/1941/31695395628_f546773ff7_o.jpg

https://farm2.staticflickr.com/1977/31695395278_c35cf8e28a_o.jpg

https://farm2.staticflickr.com/1950/31695394898_e43e0c2538_o.jpg

https://farm2.staticflickr.com/1915/31695394518_e569454541_o.jpg

https://farm2.staticflickr.com/1931/31695394198_615e03a127_o.jpg

[ToTo Man]

2015 Ditton 66 Refurbishment Photo Diary:

https://farm2.staticflickr.com/1944/45567784561_961f5807d0_o.jpg

https://farm2.staticflickr.com/1961/45567784391_2be26ef788_o.jpg

https://farm2.staticflickr.com/1967/45567784201_f32d5c5fc7_o.jpg

https://farm2.staticflickr.com/1965/45567784031_eb21a2f9b5_o.jpg

https://farm2.staticflickr.com/1903/45567783961_6ae20b017c_o.jpg

https://farm2.staticflickr.com/1941/45567783891_4e5288b6d1_o.jpg

https://farm2.staticflickr.com/1961/45567783771_d53702c443_o.jpg

https://farm2.staticflickr.com/1944/45567783581_8464af5874_o.jpg

https://farm2.staticflickr.com/1904/45567783391_bd0fa2eb54_o.jpg

https://farm2.staticflickr.com/1958/45567783251_5bc1089538_o.jpg

[karma67]

top quality work there mate,1 thing ive always wondered,is there any difference between the bass drivers with the larger dust caps compared to the smaller early ones?

[ToTo Man]

top quality work there mate,1 thing ive always wondered,is there any difference between the bass drivers with the larger dust caps compared to the smaller early ones?

I measured three pairs of T1600 and two pairs of T2169 drivers (results here: https://www.flickr.com/photos/131651119@N05/43444685622/in/album-72157699207168765/). Apart from small differences in sensitivity which are likely due to inevitable intersample variations, they all have fairly similar frequency responses. I think the T2619 version looks more impressive from a cosmetic perspective, the shallow being that I am!

[Pharos]

When I first heard these in the 70s, I thought they were better then my Tannoy Golds.

Looking at the Xovers I would now be concerned about potential transformer coupling between the coils.

[ToTo Man]

I'm looking for information on the Celestion MF500 and MD500 domed midrange units, specifically what the difference is between them?

AFAIK, Celestion first introduced the MF500 sealed dome mid unit with 50W power handling and later replaced it with the MD500 with increased 80W power handling.

Anecdotal reporting suggests the two units had slightly different frequency responses, the latter MD500 apparently had stronger mid frequencies but rolled off sooner at the bottom and top (perhaps this is the reason it was able to achieve a higher power handling).

Having measured a selection of MF500 and MD500 units (in free-space without a baffle), these are my findings:

1) The response is nowhere near flat/linear, it is around 7dB louder at 1kHz than it is at 3kHz-5kHz.
2) Left/right matching between all pairs of MF500 and MD500 units is excellent (within 1dB across most frequencies).
3) One pair of MF500 and one pair of MD500 units have identical frequency responses.
4) Two pairs of MF500 have different frequency responses between 3kHz and 5kHz (SEE GRAPH BELOW).
5) There is not a significant difference in lower/upper frequency roll-off between the MF500 and MD500 until it reaches 14kHz (SEE GRAPH BELOW).

https://live.staticflickr.com/7866/46869261804_25b246149b_o.jpg

I appreciate my sample size is small, and that measurements can be notoriously unreliable out of context, but finding #3 contradicts the anecdotal reporting that the MD500 has a different tuning to the MF500.

Possible explanations?
- Perhaps there was a there was a transition period during which Celestion mis-labelled MD500 units as MF500.
- Perhaps there was originally inherent and unavoidable manufacturing variation in the tuning of MF500/MD500 units.
- Perhaps the ageing process / deterioration is responsible for the apparently different tunings between the MF500 and MD500.
- Perhaps Celestion made a different tuning of MF500 for other loudspeaker makers (the pairs of MF500 I tested that displayed weaker output between 3kHz and 5kHz were taken from B&O Beovox speakers).

Interestingly, once installed into a Ditton 66 loudspeaker, I find the differences in tuning in the upper mid frequencies pale into insignificance in the frequency response measurements of the complete loudspeaker. This could be due to:
1) Overlap with the tweeter. The crossover frequency is 5kHz so the tweeter's response likely still makes a significant contribution between 3kHz and 5kHz. (I have measured a total of thirteen HF2000 units and all exhibit a rising response below 4kHz, though the sharpness of this rise varies greatly from unit to unit).
2) The fact that 30uf capacitance is used in the MF500 circuit vs 24uf in the MD500 circuit. (I've just learned that reducing the capacitance from 30uf to 24uf shifts the crossover frequency from 500Hz to 600Hz, so perhaps this is responsible for the MD500's increased power handling?)
3) Baffle-step effect might mean the raw frequency response measurements I made of the MF500 and MD500 in free space are irrelevant.
4) A combination of all of the above!

Are there any other plausible explanations I've missed?

It would be great to step into a time machine to find out how an MF500 and MD500 measured when it left the production line in the mid-1970's!

[ToTo Man]

I ought to change the ‘two’ in the title to ‘three’, having just completed my third - and what my dad hopes will be my last! - Ditton 66 refurbishment (…time will tell!…)

I’ve actually learned more during this refurbishment than either of my previous two, having finally taken the opportunity to measure all of my spare MF500/MD500 and HF2000 units as part of a complete 66 system. (The crocodile leads I bought to enable the speedy swapping of drive units came in REALLY handy!)

Most surprising was the extent to which the HF2000 interacts with the upper (i.e. >1.5kHz) response of the MF500/MD500, and likewise the extent to which the MF500/MD500 interacts with the lower (i.e. <10kHz) response of the HF2000. It has now become obvious to me that pair-matching by only considering the driver’s responses in isolation does not produce the expected result. Some HF2000s that measured ‘abnormally’ with one pair of MF500 drivers measured ‘normally’ with another pair, and vice versa, so lots of swapping was required to identify the best integrations! Notice how differently 'HF7' (highlighted in red) measures when partnered with an MF500 that rolls off earlier compared to an MF500 that rolls-off later).

https://live.staticflickr.com/65535/47971688908_597bb5582f_b.jpg
https://live.staticflickr.com/65535/47971679197_1576bd49ab_b.jpg

https://live.staticflickr.com/65535/47971736936_f86ca84f61_b.jpg
https://live.staticflickr.com/65535/47971677382_62b6c065b8_b.jpg
https://live.staticflickr.com/65535/47971677442_5c5b58c505_b.jpg
https://live.staticflickr.com/65535/47971677497_bb521a9215_b.jpg
https://live.staticflickr.com/65535/47971737171_8897786e1a_b.jpg
https://live.staticflickr.com/65535/47971677637_67a30b4361_b.jpg
https://live.staticflickr.com/65535/47971689458_de668d7600_b.jpg
https://live.staticflickr.com/65535/47971689953_ea27a44b5e_b.jpg
https://live.staticflickr.com/65535/47971689528_228600292b_b.jpg
https://live.staticflickr.com/65535/47971677822_cee5ee847d_b.jpg
https://live.staticflickr.com/65535/47971677912_8e5597f111_b.jpg
https://live.staticflickr.com/65535/47971689703_df3a0e3c47_b.jpg
https://live.staticflickr.com/65535/47971678047_5776daaf1b_b.jpg
https://live.staticflickr.com/65535/47971689873_76cfb9d321_b.jpg


Through lots of trial and error I settled on a driver combination for this latest refurbishment that track within about 1.5dB of each other throughout most of the audible range and have impeccably clean distortion plots, a result I am very pleased with. They were delivered to the new owner yesterday with freshly painted and polished enclosures.

https://live.staticflickr.com/65535/47985192003_1887096b5e_b.jpg

During my testing, I tried several T1600 bass drivers, several MF500 units and about a dozen HF2000 units. I noticed that regardless of the T1600 and MF500 combination used, this pair of 66s sounded consistently less coloured through the lower midrange and upper bass frequencies than my 2018 refurbished 66s. As well as being rather jealous I am also curious as to why this is?!

My 2018 66s have measurably more lower midrange and bass output than the 2019 66s, but we’re only talking a difference of about 1.5dB, which I wouldn’t have thought would have much of an audible impact on the perceived clarity and colouration of the presentation.

If I apply a low-shelf filter EQ to my 2018 66s to reduce output by about -2dB below 500Hz, their overall presentation does become clearer and less wooly/bloated and they sound much more like the 2019 66s, however they still don’t seem to have as much freedom/openness/dynamics, the kind of qualities that aren’t being picked up by the microphone. There are several potential explanations I have identified:
1) Differences in bass drivers (the 2018 66 have T2169 drivers fitted while the 2019 66 have T1600 drivers fitted).
2) Differences in mid drivers (the 2018 66 have MD500 drivers fitted while the 2019 66 have MF500 drivers fitted. I did try a lone spare MD500 driver in the 2019 66 and it sounded sublime. As with all the Ditton drivers however, each pair of MF500/MD500 was assembled by hand so will inevitably vary in their tonal and impedance characteristics).
3) Crossover differences (my 2018 Ditton crossovers were re-capped while the 2019 Ditton crossovers were left as original).

Interestingly, while testing drivers for the 2019 66, I stumbled upon one particular combination that produced the best (i.e. clearest, smoothest, most transparent and dynamic) sound I have ever heard from a 66. I literally spent the entire night listening to it utterly captivated! This was achieved by pairing my only spare working MD500 driver with a flat-measuring HF2000 unit. This MD500 in particular has approximately +2dB stronger output from 1kHz to 7kHz than any of my other MF500/MD500 units.

https://live.staticflickr.com/65535/33910246318_ca3c945d37_b.jpg
https://live.staticflickr.com/65535/33910246438_3e37f5bf84_b.jpg

Again, there isn’t a great deal of measurable difference between this combination and my 2018 66, yet the difference was clearly audible to my ears.

https://live.staticflickr.com/65535/47986000202_e9c464b8a5_b.jpg

Sadly I only have one MD500 unit with this response, its matching partner developed a buzz many years ago and regretfully I sent it away to a certain used hifi dealer in Bournemouth who was confident his tech could fix it, but when I got it back it had lost 3dB to 4dB of efficiency and had even more distortion than before!

If I wish to make my 2018 66 sound more like the 2019 66, I presume I should begin with a process of elimination starting with the low-hanging fruit? i.e. try swapping out the T2169 bass units for T1600 units, then try swapping out the MD500s? Or do you think the issue is more likely to be in the crossover? I have just been advised on another forum to remove the 1uf poly bypass cap from the MF filter because it causes a “transient pulse error” that can smear the sound. Does anyone have thoughts on this? Or thoughts on my choice of capacitors in general? To preserve the original 66 tuning I was originally advised to use ALCAPs in the LF and MF shunt positions, Mundorf ECap Plain in the MF filter, and Ansar SuperSound poly in the HF filter.

[Pharos]

I've always had a sneaking admiration for this speaker, and you've clearly got to grips with high res computer measurement, a field I am afraid of, lovely resolution.

Still I would like tidier Xovers, less spaced components, less potential coil coupling and thicker connecting wire, but this is instinctual.

[ToTo Man]

2019 Ditton 66 Refurbishment Photo Diary:

https://live.staticflickr.com/65535/47999617956_b43555c8b7_b.jpg

https://live.staticflickr.com/65535/47999617851_c1d82d2757_b.jpg

https://live.staticflickr.com/65535/47999578302_eb49d47b99_b.jpg

https://live.staticflickr.com/65535/47999618221_17bd8f8d04_b.jpg

https://live.staticflickr.com/65535/47999578812_b286368db3_b.jpg

https://live.staticflickr.com/65535/47999618181_6533f2ab99_b.jpg

https://live.staticflickr.com/65535/47999580963_5e6fcf3035_b.jpg

https://live.staticflickr.com/65535/47999581053_ff07919e6b_b.jpg

https://live.staticflickr.com/65535/47999580118_aa39f3ac62_b.jpg

https://live.staticflickr.com/65535/47999580603_fdab025591_b.jpg

https://live.staticflickr.com/65535/47999618081_ab5a403c41_b.jpg

https://live.staticflickr.com/65535/47999580518_942c421c3d_b.jpg

[Marco]

Superb job there, Richard! :clap::thumbsup:

They look like new... How are they sounding? I'm a big fan of 66s, as I had a pair myself and very much enjoyed their style of music making!:cool:

Marco.

[JohnJo]

Superb job there, Richard! :clap::thumbsup
Marco.

+1. Look fantastic. Great pictures too, thanks for taking the time to post.

[ToTo Man]

Superb job there, Richard! :clap::thumbsup:

They look like new... How are they sounding? I'm a big fan of 66s, as I had a pair myself and very much enjoyed their style of music making!:cool:

Marco.


+1. Look fantastic. Great pictures too, thanks for taking the time to post.


Cheers, guys! They sounded fantastic. I was actually sad and somewhat reluctant to let them go, but I had already committed to the sale before I'd started the refurb, - I'll know better next time! TBH I wasn't expecting them to sound as good as they did, especially with the original stock caps still in place. This suggests there is something 'magic' about the original caps that is difficult to replicate with modern equivalents.

[ToTo Man]

I decided to replace the T2169 woofers in my 2018 66s with earlier T1600 woofers, and this has brought the frequency response below 500Hz measurably and audibly closer to that of my 2015 66s. The differences that remain between 600Hz-900Hz must therefore be entirely mid unit related.

Before:
https://live.staticflickr.com/65535/48065950856_abf1a40a6e_b.jpg

After:
https://live.staticflickr.com/65535/48066055477_036faf2097_b.jpg

Out of sheer curiosity, I then decided to reverse the polarity of one 2018 66 woofer to see what effect this had:

Amplitude frequency responses:
https://live.staticflickr.com/65535/48073568251_19f8720150_b.jpg

Phase responses:
https://live.staticflickr.com/65535/48073613723_8ca1c111b7_b.jpg
https://live.staticflickr.com/65535/48073568156_db46ac133d_b.jpg

Step responses:
https://live.staticflickr.com/65535/48073613683_f00651eb16_b.jpg
https://live.staticflickr.com/65535/48073568086_ed1fa376ca_b.jpg

The difference in the amplitude frequency response when woofer polarity is inverted is pretty self-explanatory; there is decreased output between 300Hz-550Hz and increased output between 700Hz-900Hz. The mid-range sounds more forward with less upper-bass bleed when the woofer's polarity is inverted; - I think I prefer this new tonal balance.

Regarding the Phase Response and Step Response graphs, I could use a little help to explain the changes in those!

[ToTo Man]

The effect on inverting the polarity of the woofer with respect to the mid driver is even more significant when measured at farfield distance from my listening seat. Instead of elevated output between 300Hz-600Hz, there is now recessed output between 300Hz-600Hz. I think the recessed output is the more 'correct' outcome, because other speakers I have measured in my room usually exhibit recessed output between 200Hz-500Hz, so I think it's a characteristic of my listening room (e.g. floor bounce cancellation). The recessed output definitely gives a cleaner and more transparent presentation (to my ears).

https://live.staticflickr.com/65535/48085304108_6feee2b7ff_o.jpg

[ToTo Man]

Thought I should update this thread with how my Ditton 66s have been tweaked since my last post. In short, nothing much has changed. I still have the polarity of the woofers inverted to provide a cleaner response through the LF/MF crossover area. I have however since taken time to integrate my subs. I'm running the 66s full range and have set the XO frequency of the subs to where the <40Hz roll-off is minimised but the >40Hz peak isn't added to. Getting this dialled in exactly was finicky but doable with Periodic Pink Noise, an RTA and a steady hand!

I have EQ'd the entire system in 4 areas:
1) notch filter at 48Hz to capture the 42Hz, 45Hz and 52Hz room modes;
2) notch filter at 130Hz to control a response peak;
3) wide notch filter at 750Hz to attenuate mid frequencies;
4) notch filter at 6kHz to smooth MF/HF crossover region.

I used measurements to guide me where to set the centre frequency for each notch filter, but I ultimately used my ears to determine what areas of the frequency response needed to be ameliorated and how much EQ should be used to achieve this. My approach was to use the least amount of EQ possible to obtain an even-handed sonic presentation that is detailed and transparent but has no areas that draw undue attention.

The BBC-esque dip at 3kHz is a classic feature of the 66 tuning, as is the 5kHz-6kHz bump. (The 5kHz-6kHz bump is most noticeable when listening on-axis with the mid unit, which is how I prefer to listen to these speakers). I found that EQ'ing the bump completely flat robbed too much excitement so I left a little lift in.

I verified all EQ adjustments by listening and measuring in several locations around my listening seat to ensure I ended up with a response that was balanced and didn't create more problems than it fixed. Sadly I didn't save all of those measurements so am unable to upload an average of them.

The following graph shows a before & after EQ from my listening seat, with the left & right speakers averaged into a single measurement:
https://live.staticflickr.com/65535/48803313788_3ca8fefe63_o.jpg

Some will no doubt argue I've ended up with a tonal balance on the brighter side of neutral, and they are perfectly entitled to this opinion! As I say, I have not been led by target house curves, I have EQ'd the system so that it sounds both natural and neutral to my ears. :)

This was all done with the speaker grilles OFF. As much as I think these speakers look best with the grilles on, they do have a significant effect on the frequency response, knocking an average of 2dB off everything above 4kHz (the effect isn't entirely linear so I suspect the metal grid structure diffracts certain frequencies more than others), which is detrimental to the transparency/clarity and crispness of the presentation IMO.

Although it took some fettling to get there, these Ditton 66s in the current configuration are probably the most satisfyingly balanced and articulate 'big monitors' I've thus far had the privilege of owning. :)

[Pharos]

Impressive work; I wish I had the time, knowledge, and application to do similar.

[ToTo Man]

Impressive work; I wish I had the time, knowledge, and application to do similar.
Cheers, Dennis! It's a relief when the hard work pays off. There's nothing worse than putting blood, sweat and tears into something that doesn't live up to expectation, so I'm relieved I finally cracked how to make this particular pair of 66's sing.

[Pharos]

I should really have gone into the computer measurement of my own design, for which the Xover values were a guesstimate based on numerous calculations, but with no external I?P, and scarce educational resources I was intimidated.

This was also compounded by several people stating that many of the programs are not really very good.

[ToTo Man]

I should really have gone into the computer measurement of my own design, for which the Xover values were a guesstimate based on numerous calculations, but with no external I?P, and scarce educational resources I was intimidated.

This was also compounded by several people stating that many of the programs are not really very good.

I should confess that I have no EE training and was very much learning as I completed these projects. I found RoomEQWizard an indispensable aid to help corroborate and explain what my ears were hearing and I would have really struggled if I had to rely on just my ears alone. I didn't delve into the software's advanced features, I completed the projects solely based on frequency response and distortion measurements. The 'proper' way to do it would have been to model each driver's impedance curve, as this would likely have provided better insight into how they would behave around the crossover frequency. I instead took the 'suck it and see' approach of trying various driver combinations until I ended up with a frequency response at the crossover regions that my ears agreed with. It would have been much easier if I'd been dealing with modern 'cookie cutter' drive units that have far smaller intersample variations than 40+ year old hand-assembled units!

[Macca]

The following graph shows a before & after EQ from my listening seat, with the left & right speakers averaged into a single measurement:
https://live.staticflickr.com/65535/48803313788_3ca8fefe63_o.jpg

Some will no doubt argue I've ended up with a tonal balance on the brighter side of neutral, and they are perfectly entitled to this opinion! As I say, I have not been led by target house curves, I have EQ'd the system so that it sounds both natural and neutral to my ears. :)



Seems like you've got between 5-10 dB of lift in the bass, I'd have thought that would be enough to stop it sounding bright or a bit thin.

[Pharos]

I should confess that I have no EE training and was very much learning as I completed these projects. I found RoomEQWizard an indispensable aid to help corroborate and explain what my ears were hearing and I would have really struggled if I had to rely on just my ears alone. I didn't delve into the software's advanced features, I completed the projects solely based on frequency response and distortion measurements. The 'proper' way to do it would have been to model each driver's impedance curve, as this would likely have provided better insight into how they would behave around the crossover frequency. I instead took the 'suck it and see' approach of trying various driver combinations until I ended up with a frequency response at the crossover regions that my ears agreed with. It would have been much easier if I'd been dealing with modern 'cookie cutter' drive units that have far smaller intersample variations than 40+ year old hand-assembled units!

What you have done seems to me to be quite an achievement without any formal training, I have had none in computers or computing, and my Win 10 PC is such an awkward and malfunctional thing that I am dissuaded from pursuing computer measurement, but I'm sure that it would be a major aid. If you saw the pages of tables of calculation I did, to get a value for a series L for the mid/woofer, you would probably laugh. But the Xover seems inaudible, can't hear a peak or dip.
My approach is to 'nudge' the curves up to each other with small changes in Xover values, but mine was only two way, and three is much harder to do.

If you measured distortion, you must have some pretty sophisticated equipment for that.

Despite years of training my ears, I think it is very difficult to actually correlate measurements with what one hears. We used at the Beeb to train ourselves to recognise a frequency with spot checks, ie., use a sig genny and choose a frequency, and ask a 'blind' subject to say what that frequency was. After a while we got close, and we also used to see what the minimum level change that we could detect was, the 'wisdom' at that time was 1dB at 1KHz, but I proved I could detect 1/4dB. Of course my ears now at 71 are showing deterioration - partially caused by years of listening to loud music no doubt.

The ear/brain is a very sensitive device, which is why we all hear differences and agonise over them, but when it comes to assessing what is going on in a speaker by ear, even knowing the appropriate band to look at when trying to correct errors is hard. A few years ago in a discussion with Bill Woodman of ATC, I asserted that most cymbal energy was from tweeters, to which he replied that it was actually in the mid range.

I don't know what you mean by 'model' the impedance curves, just seeing the graphs would be of major help.
I agree about the units, KEF were the first to really use computer measurement to control consistency.

There are those Martin whoh think that there should be a gradual slope downwards towards the top end

[Macca]

There are those Martin whoh think that there should be a gradual slope downwards towards the top end

Indeed and looking at the FR graph that's pretty much what Richard's done.

[ToTo Man]

Seems like you've got between 5-10 dB of lift in the bass, I'd have thought that would be enough to stop it sounding bright or a bit thin.

That's a good point, Martin, but I was thinking more about the balance I have between the upper bass frequencies, mids, and high frequencies, which seems to have very little downward slope.

In recent times folk have been recommending a -10dB differential to provide a natural tonal balance for domestic listening, i.e. a consistent roll-off as frequency increases so that the highest frequencies are -10dB down from the lowest frequencies.

I can see at least one significant limitation to this: where you set the start and end points for such a target slope. e.g. A system that only has a bandwidth of 50Hz to 20kHz will sound duller with a -10dB target slope than one that has a bandwidth of 20Hz to 20kHz. My 50Hz to 20kHz bandwidth has a slope of -5dB, but the expanded 20Hz to 20kHz bandwidth has a slope of -10dB. Also, I have a broadish dip in my response between 80Hz-300Hz which could be argued to contribute to an unnatural leanness in the upper bass region, though I personally don't mind such a response (in fact in lieu of a linear response I find a dip in this area much preferable to an elevation!).

Older research by Bruel & Kjaer suggests a slope between 20Hz and 20kHz of -6dB to be optimal (see Fig.5 on page 4 (https://www.bksv.com/media/doc/17-197.pdf)).

I reckon a slope that's between -6dB and -10dB from bass to treble without excessive scribbling in between should provide a natural tonal balance, but at the end of the day it also comes down to what the individual listener prefers.

I should note that my measured response was with the grilles off. Putting the grilles on the speakers knocks an average of -2dB off everything above 4kHz, which tips the presentation into sounding too dull/veiled IMO. The grille also affects certain frequencies more than others (it has a metal grid structure which I suspect causes considerable diffraction), another reason why I listen with the grilles off. It's a shame as I think these speakers look better with the grilles on, and they also help reduce my anxiety over dust getting stuck to the tacky surface of the MD500 dome, which would be a PITA to remove!

[Macca]

I agree one size is never going to fit all. There's personal preference in terms of how toppy or bassy you want it.

[ToTo Man]

What you have done seems to me to be quite an achievement without any formal training, I have had none in computers or computing, and my Win 10 PC is such an awkward and malfunctional thing that I am dissuaded from pursuing computer measurement, but I'm sure that it would be a major aid. If you saw the pages of tables of calculation I did, to get a value for a series L for the mid/woofer, you would probably laugh. But the Xover seems inaudible, can't hear a peak or dip.
My approach is to 'nudge' the curves up to each other with small changes in Xover values, but mine was only two way, and three is much harder to do.
I can't help you with Windows I'm afraid, Dennis, having been a lifelong Mac user. The last time I used a PC, Windows XP was still the most widely used OS!


If you measured distortion, you must have some pretty sophisticated equipment for that.
Quite the opposite actually. All I use is a miniDSP UMIK-1 USB microphone and RoomEQ Wizard software. The software displays a plethora of graphical data from a single frequency response sweep. I don't know how accurate/reliable the distortion data it produces is, but it certainly helped me to weed out mid and high frequency drivers that had abnormal resonances.


Despite years of training my ears, I think it is very difficult to actually correlate measurements with what one hears. We used at the Beeb to train ourselves to recognise a frequency with spot checks, ie., use a sig genny and choose a frequency, and ask a 'blind' subject to say what that frequency was. After a while we got close, and we also used to see what the minimum level change that we could detect was, the 'wisdom' at that time was 1dB at 1KHz, but I proved I could detect 1/4dB. Of course my ears now at 71 are showing deterioration - partially caused by years of listening to loud music no doubt.

The ear/brain is a very sensitive device, which is why we all hear differences and agonise over them, but when it comes to assessing what is going on in a speaker by ear, even knowing the appropriate band to look at when trying to correct errors is hard. A few years ago in a discussion with Bill Woodman of ATC, I asserted that most cymbal energy was from tweeters, to which he replied that it was actually in the mid range.
The human ear is indeed an incredibly sensitive device and is sensitive to changes that are often barely detectable in measurements, but I also agree that it is very easy to confuse/doubt oneself as to what it is that has changed, which is why measurements can be helpful in providing direction and/or validation.


I don't know what you mean by 'model' the impedance curves, just seeing the graphs would be of major help.
That's probably because I've used incorrect terminology! I mean measuring the impedance curves, using a system like DATS (http://www.daytonaudio.com/index.php/dats-dayton-audio-test-system.html). Apparently, due to the ageing process and hand-made nature of the Celestion drive units there can be significant intersample impedance curve variations between otherwise seemingly identical units, variations that only manifest as significant frequency response differences when the drivers are actually connected to the crossover (when measured in isolation and without a crossover they often have near identical frequency responses, which can be misleading).

[Qwin]

Very nice project/s.

[ToTo Man]

Very nice project/s.
Thanks Ken, I'd have really struggled with them if it wasn't for your patient answering my endless questions! :)

[Pharos]

The downward slope idea is not an established and universal criterion. It is touted by some designers, but many speakers are still designed to be, and quoted as, flat, mine are 22Hz to 50kHZ +/- 3dB, so must be flat. (with wiggles)

I was a Mac user from'91 onwards, SE30, (which died in '05) but as the price increased and I became poorer I adopted the wretched PC school, going through 4 PCs and '95, 98, XP, Win7 and now 10, I think XP was the most stable and sensible.

I've just bought a second hand Mac running Ojave. Lovely, but I am not yet very familiar, and I tend to 'save it for best'.

Yes in those early days driver irregularities were often offset with tweeked Xovers.

[Macca]

The downward slope idea is not an established and universal criterion. It is touted by some designers, but many speakers are still designed to be, and quoted as, flat, mine are 22Hz to 50kHZ +/- 3dB, so must be flat. (with wiggles)




Most speakers are designed to be flat anechoically, so that once you put them in a typical room you get boundary gain and HF gets soaked up so you get a tilted response downwards from bottom to top. If they measure flat in a typical room they'll probably sound a bit toppy. I do wonder about the people who find modern speakers too bright and how sparsely furnished their rooms are - wood floors, hardly any furniture, nothing on the walls and so on. Might be the reason.

[Pharos]

Alan Shaw has posted much about how he thinks many current speakers are toppy.

One of my main concerns ATM is that, I certainly, and probably all of us, become conditioned to a particular character of sound, and then everything else can sound wrong in comparison

[ToTo Man]

Alan Shaw has posted much about how he thinks many current speakers are toppy.

One of my main concerns ATM is that, I certainly, and probably all of us, become conditioned to a particular character of sound, and then everything else can sound wrong in comparison

I've always suffered from this issue, - my ears get accustomed to the tonal balance of the speaker I'm listening to, then when I switch to one of my (many!) other speakers it's quiet jarring until my ears become acclimatise to that speaker's presentation. And then when I revert back to the other speaker it's that one that now sounds 'wrong'. It's fascinating the tricks one's brain can play on oneself!

I do agree that modern living has a lot to answer for when it comes to messing up the acoustics of a domestic hifi system. I am bemused by those who spend inordinate sums on boxes and cables without giving a second thought to their listening environment. I had my room kitted out in bass traps and first reflection panels in 2013 and could not contemplate going back to listening to music without them.

[Macca]

Some people are stuck with a communal room and don't have the option of using bass traps or other treatment. And if the wife is wearing the trousers then they probably can't put in carpet, rugs, big sofas or anything else that might help unless the aesthetic is to her taste. No wonder the speakers sound bright.

the other reason could be that modern speakers over a certain price point are pretty good i.e they have ruthless clarity. So the problems with mediocre sources and amplification are shown up. It isn't brightness they are hearing, it's distortion.

[ToTo Man]

Time to resurrect this thread once more!

After experimenting with in excess of twenty HF2000 tweeters, I have now settled upon a matched pair for my original 2015 refurbished 66s. I suspect this will be as a good as it gets when it comes to high-frequency pair matching.

1-metre FR before:
https://live.staticflickr.com/65535/50301817581_123e6505ae_o.jpg

1-metre FR after:
https://live.staticflickr.com/65535/50301969567_7e0993e930_o.jpg

This is how the new driver combo measures from the listening seat before & after EQ:

Listening seat FR before EQ:
https://live.staticflickr.com/65535/50301817481_c4689c57a0_o.jpg

Listening seat FR after EQ:
https://live.staticflickr.com/65535/50301969592_94265ef5dd_o.jpg

Listening seat average FR before & after EQ:
https://live.staticflickr.com/65535/50301969512_b6f7d1334b_o.jpg

Was it worth the effort? I'm hardly going to say no, am I??!! The latest tweeters have smoother FR and less output than the previous ones and to my ears integrate better with the mids and draw less attention to themselves. Not a night and day difference, but still worth the hours spent listening to and measuring numerous pairings. Imaging has become even more stable and holographic as a result.

I've also changed how I listen to these speakers. The plywood plinths have been temporarily ditched and I now have them sitting directly on the carpet but tilted backwards. I've discovered this provides exactly the same benefits to midrange-treble cohesion as elevating the enclosures on 25cm plinths, but with the added benefit of making it FAR easier to move the speakers out of the way when I want to listen to my Tannoys! :) Ideally I'd love my Tannoys to be on castors so that I could wheel them out of the room and place the Dittons closer to the wall, as the bass sounds a little anaemic to my ears (I suspect it's being adversely affected by SBIR, but that's a topic for another day!...).

https://live.staticflickr.com/65535/50302216687_d96a601e04_o.jpg