Dynaudio identification help required.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Last edited:
Well, there's more than one way to skin this cat (sorry animal lovers), and the ICE powered DSP plate amp would certainly deliver far more system flexibility than the original, and the elegance and simplicity of no external components.

Thanks to the modern offerings of very compact class D amp boards / modules, the simpler approach of internal amp and passive XO would likely be a less costly approach, but would lack the advantages of full DSP processing, including EQ, etc

So I'd imagine the path forward depends on whether the cited miniDSP unit would fit without much modification, and if owner wants to invest in the steeper cost. If it weren't for the software's learning curve, I might have tried that already myself :rolleyes:
 
Owner managed to picked up a mf a300, so were going the passive route.


636816d1506133623-dynaudio-identification-help-required-front-jpg

Too nice to throw away, for sure.

Looking at the scorch marks, the electronics was insufficiently ventilated. Sort of issue you'd run into with a near prototype.

With the low price of second-hand amplifiers in junk shops, I'd think that going passive is a very real possibility. My own view, is that if you design for easy impedance, you really don't need any exotic amplifier for a very good sound.

And you do that by using an old BBC BW3 design like the LS5/9:
Rogers Loudspeakers › LS5/9

This speaker is still made, and still sounds good. It uses passive RL bafflestep and impedance correction: mh-audio.nl - Home

Just calculate for BW3 and 10R bass and 6R tweeter and 3500Hz, and the values drop out: 2-Way Crossover Calculator / Designer

But it's down to your friend really. FWIW, I believe the D28-2 is a ferrofluid design, so an oil-change might be in order.


Steve, just out of curiosity, why are you suggesting a 3.5khz xover point ?
With an 8" bass driver, wouldn't a lower point be preferable ?



I think I have seen this speaker in a special "speaker-issue" of the German DIY magazine "elrad".
I know that I still have all these mags but I don't know where at the moment.
Regards
Charles


Hi Charles, if you could find and post that article, that would be so awesome.
I had a look around for that magazine and article but couldn't find it.
 
The reason that a highish 3.5 kHz crossover works in two-ways with big, badly time-aligned woofers is something the BBC discovered 50 years ago. :D

Rogers Loudspeakers › LS5/9

See with a big woofer, your time-alignment problems to the tweeter come down to about 5cms.

Well, at about 3.5kHz, it happens that the speed of sound, being about 340m/s, to be about 5cms as a half-wavelength. :cool:

Hence these negative polarity classic BBC designs: Rogers Loudspeakers › LS5/9.

I'm all in favour of good amplifiers: Musical Fidelity A300 Integrated Amplifiers reviews - Audioreview.com

BW3 and impedance correction is quite a hard concept to explain. It happens to be somewhere in-between LR2 and LR4. And therefore viable IMO.

What I can tell you is that the BW3 concept, and recessing the woofer, and flat impedance is still alive and well. Some people like it. I do. :cool:
 

Attachments

  • Modified_MA-R300MD_HT22-8.jpg
    Modified_MA-R300MD_HT22-8.jpg
    79 KB · Views: 171
  • Wharfedale_Melton_2.JPG
    Wharfedale_Melton_2.JPG
    47.1 KB · Views: 159
  • Harbeth M30.1.JPG
    Harbeth M30.1.JPG
    33.2 KB · Views: 160
Last edited:
Thanks Steve.
Time for me to go do more research :)

Tweeters done, woofers done, now I need to sort out the cabinets, then do some sweeps...

Why do the "it's a simple job" repairs end up a pain in the @@@@@ :)
Oh yes, cause no one would buy the thing if it was advertised correctly.
Dog looking for cashed up new owner...
 
The crossover frequency with such an unusual 2 way set ought to be on the lower side. Tweeter is capable of performing nicely at 2 kHz, so in that region it would be wiser to high pass it. A good compromise between tweeter power handling and off-axis response of the woofer. If you ask me.
 
An update for those interested, and in the interest of documenting history.

I'm still working on them, very slowly, as time and other jobs permit.

The people at Dynaudio have been able to provide some more information about these speakers.
Heres a selected copy and paste of Dynaudios email to me.
....

A gent by the name of Wilfried remembers this old model.
It was a DIY model called Jadee Active.
Dynaudio used to sell drivers and kits for DIY speaker builders but this part of the business was discontinued long time ago, around 1996 according to Wilfried.
Many of these kits are still alive out there and we frequently get requests for spare parts for these.
Unfortunately they are now so old that this is often not possible - but we always try!
The tools for the old drivers are all but gone and technology has moved quite a bit since the 1990's.
....

This is all the info that they have been able to provide.
But it's another step forward :)

Huge thanks to all the people at Dynaudio for their time and help.
 
It's always nice when people get back on projects. 'Cause I am interested.

I did a bit of work on this today. My great inspiration on 8" plus 1" is Troels Gravesen: DTQWT TWEAKS

Now, TBH, I figured out most of this approach on my own... because I build speakers like this. o

I simmed it time-aligned on a typical 1" 88dB tweeter. And any old 8" bass. I set the tweeter back 5cms. Which flipped the polarity to all positive.

I think these sort of circuits are worth a go. The ca. 6kHz tank notch ( I'd try 0.47uF or 0.68uF with 22R) is interesting. It certainly does something to take down the cone breakup from the bass and clean up the sound. Tweeter level, as always, is adjustable to taste.
 

Attachments

  • Dyanaudio Sim Circuit 3rd order.PNG
    Dyanaudio Sim Circuit 3rd order.PNG
    6.9 KB · Views: 78
  • Dynaudio Sim FR 3rd order.PNG
    Dynaudio Sim FR 3rd order.PNG
    17.5 KB · Views: 81
  • Dynaudio Sim circuit 2nd order.PNG
    Dynaudio Sim circuit 2nd order.PNG
    8.4 KB · Views: 71
  • Dynaudio sim 2nd order  FR.PNG
    Dynaudio sim 2nd order FR.PNG
    20.7 KB · Views: 85
Because of the rarity ? of these speakers, I almost class this as giving back to the audio world.
A blog or historical record if you will.
Hopefully this will help others now or in the future.

Steve, you really are a gentleman.
Giving of your time and experience and ask for nothing in return.

The owner asked for a change in the cabinets appearance to better fit the waf :)
Once the paint and varnish has cured a little longer I will mount the drivers and do some sweeps that I will post here.
Pics will also be forthcoming.
 
I am happy to see people care about restoring stuff. If anyone ever feels like purchasing a similar Dynaudio unit, there is an automotive unit that used to be installed in Volvo automobiles, and now available on ebay, if you find suitable in a nice condition. I've had one opportunity so I grabbed a pair. It plays very nice in a sealed 40 litre enclosure which I concluded from test driving it without XO. One day I may build a 2 way waveguided tweeter design with it. Initially I thought I'd have a second pair of woofers as spares for my 3 way's. Drivers are about 18 years old. The model is just a name I made up.
 

Attachments

  • Volvo 24W-R.jpg
    Volvo 24W-R.jpg
    118.7 KB · Views: 168
  • Dynaudio Volvo 9.5 inch woofer-2.JPG
    Dynaudio Volvo 9.5 inch woofer-2.JPG
    280.9 KB · Views: 218
  • Dynaudio Volvo 9.5 inch woofer-1.JPG
    Dynaudio Volvo 9.5 inch woofer-1.JPG
    289 KB · Views: 151
Lojzek - Interesting..

Finally got the drivers in late last night.
So here's some fresh pics, info, and sweeps...

The owner wanted the cabinets darker to try and help with WAF.
So I've sanded them back and used a tinted oil based varnish (one coat) then three coats of clear.
I filled a few holes on the back panel before respraying those in the original satin black.

Cabinet info; All 16mm chipboard.
Upper two sections have 'egg crate' foam on most surfaces.

Top section 44.5 litres - Internal, 260mm wide, 308mm deep at top, 365mm deep at bottom, 510mm high.
Three Variovents feed into bottom section.

Bottom section 23.2 litres - Internal, 260mm wide, 365mm deep, 245mm high.
This has two Variovents on the back panel.

Foot section 18 litres - Internal, 260mm wide, 365mm deep, 190mm high.
This is sealed of from the upper two chambers and was used to house the original power supply and amplifiers.
As the amps are beyond repair it will now hold the passive crossover.

Drivers;
D28/2 Tweeter - I've put new diaphrams in both.
24w75 Bass - I've refoamed both.

On to the pics :D

Yes the cabinets aren't perfect, but I've done what I can.
Re-veneering would be nice, but it's out of the budget.

Arta sweeps are of the bass and tweeter at 1 meter on axis. same volume. straight from the amp.
Looks like maybe I'll go ~2.5k xover, tweeter padding, and of course a zobel for the bass.
 

Attachments

  • dyn sweeps 1.jpg
    dyn sweeps 1.jpg
    244.3 KB · Views: 145
  • fresh.jpg
    fresh.jpg
    150.3 KB · Views: 260
Been an awesome day.
Managed to do so much despite the !@#$%^& usb microphone dropping out constantly.

A zobel of 18uf and 10ohm seems to give the smoothest impedance for the bass driver.

I threw together a 2nd order LR at 2.5khz, with a tweeter l-pad of 3.6ohm series and 7.5ohm parallel.
But this gave a big dip at the crossover area.
It was clearly audible too.

Then I decided to try Steve's suggestion of a 3rd order Butterworth.
This should boost the crossover area by about 3db.
Just what it needs to get a 'flat' response compared to the LR.

Again needed to put an l-pad on the tweeter. Currently using 2.7ohm series and 12ohm parallel.

But presto, with the same 2.5khz crossover point, the dip is gone and even with bottom end parts it sounds a heck of a lot better.

Later when I can be bothered ill reverse the tweeter and see what sort of 'suck out' it does.

Oh, and I'm basing my crossovers on 7ohm bass and 7ohm tweeter.

If anyone thinks I've done wrong or can do better, feel free to say so. :)
 

Attachments

  • system bw3 + 12p 2.7s.jpg
    system bw3 + 12p 2.7s.jpg
    201 KB · Views: 49
  • system lr 2.5khz 3.6s 7.5p l-pad.jpg
    system lr 2.5khz 3.6s 7.5p l-pad.jpg
    200 KB · Views: 58
  • bass 10ohm 18uf zobel dats 30-10-2017.jpg
    bass 10ohm 18uf zobel dats 30-10-2017.jpg
    235.5 KB · Views: 53
Could you be more specific as to what values that LR filter was? Have you thought of simulating in the usual way, taking measurement of both of the units, frd's and zma's, assessing Z offset, noting baffle driver coordinates...etc as it had been discussed over here many times? I am attaching my measurement of Dynaudio D28/2 that I exported into Response Modeler. It appears that your version has suspiciously high voice coil inductance, assuming you took raw driver IMP measurement. I am using Clio measurement suite.
 

Attachments

  • Dynaudio D28-2.png
    Dynaudio D28-2.png
    20.2 KB · Views: 44
Sure thing.

I used 4.7uf and 1mh.
Not exactly what the internet calculator suggested, but I thought it would be close enough to get it going.

Ive used dayton dats for impedance sweeps.

Dont forget I also replaced the tweeter diaphrams using aftermarket ones.
They are stated as being 'suitable for the d28/2'.
But we all know that might not be 100% true ;)

Ive just done a new dats sweep of the other tweeter, with a close up of the measured parameters.
Hopefully that helps you :)

I have various programs that should be able to do frd etc files.
but i havent had the time to try and learn them yet.
i do know trying to export/import from arta gives, um, mixed, results due to formating issues.
 

Attachments

  • 2nd lr 2.5khz 7ohm.jpg
    2nd lr 2.5khz 7ohm.jpg
    39.6 KB · Views: 91
  • tweeter dats free air test.jpg
    tweeter dats free air test.jpg
    230.8 KB · Views: 61
  • tweeter dats free air test panel.jpg
    tweeter dats free air test panel.jpg
    68.5 KB · Views: 58
I see, that explains it. I could help with a workable XO design if you can provide raw driver frd's and zma's at a specific mic point in space, for instance 1m from the baffle, between the units, distance from woofer centre to TW centre. You'd post exported files of woofer in the box frd, woofer in the box zma, TW in the box frd, TW in the box zma, W and TW in parallel in the box frd ( Z offset). Time window large enough to capture 100 Hz with enough of data points to represent the actual graph. As far from the obstacles as feasible. Best zipped.
 
Last edited:
I've had a look through arta, as far as I can see I can only export files as ascii or csv.

The att bass driver sweeps are measured at 1 meter, mid point between bass and tweeter centres.

And right now, thats all I can do.
The microphone has decided to fail completely.
So until I get a replacement this project has once again come to a stop..............

Frustrating.
So close.





So faaaaaaaaaaaaaaaaar away.
 

Attachments

  • Bass. 1 meter. Mid point axis. csv.txt
    15.5 KB · Views: 28
  • Bass. 1 meter. Mid point axis. ascii.txt
    14.2 KB · Views: 32
I'm not actually having a good time with this one. AFAIK, because it is time-aligned, which I simmed with 5cm recess on the tweeter, all our butterworths are becoming phase aligned, and creating a +3dB peak at crossover. And the linkwitz-rileys are showing a hole in the midrange.

I'm not too sure at all what I am getting here. I've got enough for a couple of bass light builds with nice impedance. You can ignore the bass adjustments which I did for an 8 ohm woofer and just insert your own impedance correction. Which must be around 9R and 22uF.

I found with the 2500Hz circuit, it looks better with 1mH and 10uF second order, but that mucks up the impedance. A peak at crossover isn't the end of the world. It sounds OK off-axis. So don't toe them in.
 

Attachments

  • BW3 3500Hz 8 ohm DCR Calculator.PNG
    BW3 3500Hz 8 ohm DCR Calculator.PNG
    19.4 KB · Views: 56
  • BW3 3500Hz time aligned circuit 8 ohms.PNG
    BW3 3500Hz time aligned circuit 8 ohms.PNG
    8.2 KB · Views: 49
  • BW3 3500Hz Time-aligned 8 inch plus 1 inch.PNG
    BW3 3500Hz Time-aligned 8 inch plus 1 inch.PNG
    18 KB · Views: 49
  • BW3 2500Hz theoretical.PNG
    BW3 2500Hz theoretical.PNG
    19.1 KB · Views: 219
  • BW3 circuit 2500Hz.PNG
    BW3 circuit 2500Hz.PNG
    8.5 KB · Views: 45
  • BW3 2500Hz FR.PNG
    BW3 2500Hz FR.PNG
    20.7 KB · Views: 45
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.