Refurbishing Acoustat 2123 interface, have ?'s

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Michael,

Nice to read so much and too see these pictures of Acoustats...
The sound level limit of the Acoustats is something I've been trying to improve. So I can pehraps help a bit ...

I still own a pair of Spectra22 too, and I hurd the (more than) excellent 1+1's of André last spring. Thanks again for your welcome André !
As you know is 1+1's have been widely modified, to the point that the comparison with the stock Acoustats makes almost nonsense.

My 1+1 have been also largely modified but not to the point as as yours André. I'm about building steel frames but this damned time is going too fast... I'll keep you informed.
Parallel I also own a pair of spectra 1100,
but I spent most of the 2 last years on the building of a pair of Spectra44 which I finished 6 month ago. Another galaxy.


About amplifiers :
... I burnt 3 amplifiers on the 1+1's : a Nad 218, and two lovely Carver PM1.5. So do not even try these last ones. The sound was really good, but it seems that the 'magnetic field' system does'nt appreciates these stats.
- I had good results with a Threshold 400A
- On the Spectras2200 I also tried a 40W Cary valve amp. Very nice indeed but not enough vigor to my taste the amp when quickly to a kind of 'soft distortion' when playing loud.
- Since 4 years I'm using a pair of Electrocompaniet AW180 with full satisfaction on every Acoustat models I've.

My favourite choice if I'd the monney would be a big classA amp (Plinius 103/250 or similar)

... And take great care to the source : the Acoustat will tolerate a 'medium quality' powerful amplifier if it stands the load, but they are very demanding for the preamplifier.
=> The preamp (or the DAC) shall be musical and very dynamic, able of far more than 2 volts ouput (that my experience once again).

About the 'limitations' :
Spectra 22 have been among the finest speakers of their range, and they are still in the race in terms of transparency, imaging and speed.
BUT, from my experience : do not expect very large sound levels from Spectra22. Find a pair or 2+2 or -even better- Spectra44 for that.

To be clear about what we call 'high sound levels' in a 80 square meter room :
- an AC-DC or kinda music played so loud that you cannot speak together in the room. But without any audible distortion.
- three opera singers (soprano + alto + baryton) singing at a level you can barely stand (so close to the reality), but once again without any distorsion.
That makes about 92-95db I think but the size of the room matters a lot.
Such sound levels you can obtain quite easily with large electrodynamic loudspeakers.

But ... with a pair of Spectra22, or the others '2pannels' ... no hope... the fuses will blow, then the transformers if you insist too much...
The detailed reasons are too long to explain, pehraps do you know them already : low efficiency, small displacement of the diaphragm between the electrodes, very low impedance, capacitive impedance, arcing ...
Many ingredients of a low level limit are there.

As I don't know what you have already done, there are nevertheless some possible improvements I experimented :

The 1 and 2 are the most audible.
1 - The most efficient : the resonnancy of the pannels is about 50Hz : the trick is to limit the frequency below this frequency, this will dramaticaly reduce the diapragm displacments, thus reducing the distorsion at high levels.
There is a 100Hz filter included to the stock interface, but it removes too much of the 'soul' of human voices to my taste.
Another way to do this is to add a 0.047 to 0.1uF cap between the preamp and amp (if this has a standard 47Kohm impedance). Just make the test. This gives relief to he pannels.
... Of course there is a loss of bass (typicaly below 35Hz with a 0.1uf cap). You can accept to live with, or add a good active sub.

2 - ! Weld the heads ! Connect together with a fine copper wire the top and bottom connections of each pannel. This is an old and basic (and cheap) modification, one of the most efficient. Takes time, about 6 hour/pannel.
I can send photos if you wish. Many guys in this site know this mod.

3 - Some components in the interface seem to don't like very much high power : I advise to change the stock resistance against Caddocks ones (same values). My contact : Michael Savuto of Analogue associates.
This gives many improvements, among them a more dynamic sound, thus the feeling of a higher level.

4 - Suppress the speakers cables, by using monaural amplifiers ... or by moving the interfaces transformers very close to the amplifier with very short cables (about 20cm).
=> Then use a High voltage wire to bring the HV signal from the interface to the pannel. The high voltage can travel long distance without any loss.
My contact : Rob Mcinlay of ERAudio (cheap).
This gives additional transparency, and dynamic too.

These mod's are worth the time you will spend, and will give a new life to your speakers.
But if you really need 'big sound', keep the 2200 as they are, and look for a pair of 2+2 or Spectra 44 !

Long life to the Acoustats,
Mathieu
 
Good day,

Cannot link picture to the DIYAudio site... 'not geek enough...

=> I sent to you the complete mod doc of my 1+1 per email.

Most of the mods are ok, only the modification of the HF cap value was not a good one. Stay close to the stock value (between 0.0085 and 0.0115uF) for a nice result. I recommend more modern caps than the stock yellow ones (Auricap, ASC...). That my taste.

Did you receive it ?

Mathieu
 
crossover change

The question is what would it take to change the crossover that is built in from 100hz to say 55hz? it would seem that it should be as easy as changing the values of the caps?

Any heads up on this would be of great help as I am about to replace the switches and electrolytics in the crossover. I have ZERO intent of using a stock sub on the system, since it has 2 rhythmic 12's running with them right now, but it would be nice to get rid of just a bit of the bass that the panels are trying to handle for now ( at least until I get another set of 9 inch 5 wires to make 33's)

Jason
 
I have some Spectra 22 and how do you tell the difference between the
older ones and the newer ones? I would also like to know to weld the heads.
Any Info. would be appreciated.
Oldcarsaz :cool:


Not sure what you mean by "older" and "newer". If you mean the difference between Spectra 22 and 2200, the difference is purely cosmetic, with the 2200 having wood trim with brass inlay around the top and sides of the panel. Electrically they are identical. If that doesn't answer your question, please clarify.


Please explain what you mean by "weld the heads".:confused:
 
They talk about an antenna. I don't see one. To weld the heads they say to run a thin copper wire through the top and bottom. Is that all the wires top and bottom?
Thanks
Oldcarsaz


Ahhh...you are asking about older and newer versions of the Ultrasonic Bias Power Supply. On the air-core transformer located on the small PC board, there is a red wire exiting the top of the transformer, the other end of which is attached to the main PC board. If you see another wire wrapped around the red wire (but not connected electrically) that is the "antenna" used in the improved version of the bias supply. There are other changes that go with it, but that's the way to quickly identify which you have. Keep in mind that if your supply is working reliably, there is no reason to upgrade.


I still don't understand the "welding heads" question. What are you referring to when you say "heads"? I might guess you mean soldering, rather than welding, since other than the steel case, there's nothing that was, or should be, welded in the speaker.


Perhaps you are referring to the highly questionable and risky modification of bridging the stator wires at the top and bottom of the panels. I see no point in doing this, and you certainly risk damage to the speaker in doing so. Furthermore, in a Spectra speaker, due to the segmented stators, bridging the wires is even trickier, and if done improperly, will ruin the segmentation.


If I've guessed incorrectly, please clarify what you mean.
 
I have 2 red wires coming of the board they both have LF and go down under the
transformers.
oldcarsaz


You're looking in the wrong place. There are three PC boards: one medium one, buried underneath the high voltage board. The switches, binding posts and xover components are mounted on it. Not this board.


The largest of the three is the high voltage board, which has the bias quintupler and high voltage capacitors/resistors mounted to it. The panel connections are made to the edge of this board. Not this board.


The third board has the ultrasonic power supply, and is mounted at right angles to the chassis. The AC power input jack and pilot light are mounted to it. This is the board that contains the air core transformer, usually with a black wrapper, about 2" square. Look at this transformer's single red wire to see if it has the "antenna" wire wrapped around it.


Again, there's no need to update this circuit if your bias supply is working okay. This is NOT a sonic upgrade. If you have a bias problem, I can post the upgrade instructions. The change is fairly simple using commonly available parts, but the supply must be re-calibrated afterwards, which requires a high-voltage probe for best results.
 
acoustat mk-2123 problem

i have 2 of these. one works and sounds fine and one distorts at higher volume.i just put my meter across the red and blk terminals and on the good one it reads around 4 ohms.on the distorted one it reads 2 ohms.i also just blew a channel in the receiver that was driving the distorted one. question? why is the bad unit only reading 2 ohms? im sure this is what blew my receivers channel.do i have a bad transformer or cap? Help??
 
Assuming you made your measurements with the amplifier disconnected from the speaker, and measured from red binding post to black binding post, the 2-ohm reading on the distorted speaker almost certainly means you have an internal short in one transformer (or possibly both). The transformer cannot be repaired. Both audio transformers are driven in parallel (but in anti-phase), so you you'll need to disconnect their primaries (four black wires) and compare measurements on each one. If one measures significantly lower, you've found your bad transformer. Be sure to mark each of the wires so you can return them to their respective connections. See attached schematic.
 

Attachments

  • Spectra2200-3300 Schematic.pdf
    212.2 KB · Views: 111
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.