ACE-bass amplifier design

[steph_tsf]

There is a balanced bridge application called Servo-Sound active speaker.
US Patent 3647969 - Tadeusz Korn Servo-Sound active speaker
Unfortunately, his patent is unclear about how to connect the transistor-based circuit represented as FIG. 5 inside the patent.
Tadeusz did it on purpose, I guess.
If you are lucky enough to find such Servo-Sound active speaker, you may try reverse-engineer it.
Unfortunately, all components involved into FIG.5 are potted with resin.
Some time ago somebody told me one can dissolve some potted resins using acetone or possibly tetrachlorure, I don't remember precisely.
Watch out for toxicity. Read first about acetone and tetrachlorure !

[consort_ee_um]

One way to mitigate the voice coil heating problem is to use the Dynaudio type drivers which have a massive voice coil and hence surface area to give better cooling. The Hi-Vi, Dynavox and Morel copies of the Dynaudio are available but now relatively expensive. I do not think you can buy Dynaudio drivers anymore but the complete speakers are relatively cheap, so might be a source.

[forr]

The purpose of sensing the output current (depending of course of the back-EMF of the driver) in Stahl's method is to achieve a power amplifier with a special complex output impedance such as the whole electrical system, amp+driver is deeply modified from the usual configuration of pure voltage drive.

Albeit probably not very realistic, an open-loop amp having a negative impedance and additional features as required in Stahl's method is conceivable,
the concept would nevertheless remain entirely valid.

Stahl has made it very clear in his JAES article :
"ACE Bass Is not a Feeback System."

*

Concerning the thermal behavior of the voice coil, its effect seems to be largely overestimated for loudspeakers in domestic conditions. Most people having experience with Voigt's bridge feedback and the like have found unnecessary to compensate it.

However, Yamaha deals with the problem, and Ian Hegglun proposed a nice DIY solution in "Speaker Feedback", Wireless World, May 1996, p380.

*

There is an overlooked problem with the techniques using driver current sensing : the inductive nature of the voice-coil which is constant neither with frequency nor in the voice-coil position in the gap.
It is an annoying factor which somewhat limits the achievable potential, no more 10-12 dB of correction has been mentioned in the case of Voigt's bridge.

[CharlieLaub]

Originally Posted by forr

The purpose of sensing the output current (depending of course of the back-EMF of the driver) in Stahl's method is to achieve a power amplifier with a special complex output impedance such as the whole electrical system, amp+driver is deeply modified from the usual configuration of pure voltage drive.

Not sure what your point is above...

Originally Posted by forr

Albeit probably not very realistic, an open-loop amp having a negative impedance and additional features as required in Stahl's method is conceivable,
the concept would nevertheless remain entirely valid.

Stahl has made it very clear in his JAES article :
"ACE Bass Is not a Feeback System."

That should read "ACE Bass Is not a Motional Feeback System."

ACE-BASS most definitely uses feedback around the driver!

Originally Posted by forr

Concerning the thermal behavior of the voice coil, its effect seems to be largely overestimated for loudspeakers in domestic conditions. Most people having experience with Voigt's bridge feedback and the like have found unnecessary to compensate it.

I have seen studies (not of ACE-BASS systems) that showed that voice coil heating in a loudspeaker was not significant. The driver, however, were not undergoing the massive power boost that typical ACE-BASS systems apply, and so will not undergo the same level of VC heating. You realize that ACE-BASS does boost power to extend and flatten the response, don't you? You can easily see this if you model the circuit using Basta!. The requirements for power are often quite high.

Actually there was a study, referenced earlier in this thread I think, that showed that the frequency response of ACE-BASS can vary from -2dB to +10dB with voice coil heating. Also, the "benefits" of the circuit are only obtained when Re is very closely cancelled by the negative output impedance of the amplifier.

Originally Posted by forr

However, Yamaha deals with the problem, and Ian Hegglun proposed a nice DIY solution in "Speaker Feedback", Wireless World, May 1996, p380.

*

There is an overlooked problem with the techniques using driver current sensing : the inductive nature of the voice-coil which is constant neither with frequency nor in the voice-coil position in the gap.
It is an annoying factor which somewhat limits the achievable potential, no more 10-12 dB of correction has been mentioned in the case of Voigt's bridge.

I am very interested in getting a copy of Hegglun's article. If you or anyone has one that would be willing to share, please contact me.

[bentoronto]

Can somebody explain how you can have feedback including the speaker and not consider it motional feedback?

As far at Stahl's remark is concerned, people will say the darndest things to qualify for a patent. As I (and more importantly on this topic, Elliot) said before, long history of exactly that.

With montional feedback, there is more power pumped into the VC at those very rare times when there is a music signal that low (say, below 25 Hz). But at resonances, I would guess a MF amp cuts the power in order to cut the motion. For very short moments of overshoot, I'd guess the MF amp does provide strong power to the driver in order to counter-act the over-shoot.

To assess whether heating is a big issue or not, you need to know the distribution of peaks in actual music reproduction, not test with continuous tones in a lab.

Ben

[frankz55]

I intend to build a pair of speakers with the ace bass system. seeing the patterns of post # 11 and # 20, I can't figure out how to calibrate the P4 and Pot "a" pots from 100k, if someone would tell me the procedure I thank them.
I would also like to use an Ljm L7 as an amp, is it possible?

[TNT]

Hi!!

If you have any questions to the original inventor (K-E S) you can post them here - he's my neighbour

//

[frankz55]

Originally Posted by TNT

Hi!!

If you have any questions to the original inventor (K-E S) you can post them here - he's my neighbour

//

I'm already in possession of a pair of a4-14mk2 and sub b2-50, they sound good to me.
I would like to build a similar pair for my brother, using inexpensive speakers on the market (Polish STX type).

[esl 63]

Joe has more to say here:
Joe Rasmussen "Trans-Amp" - 40 Watt Transconductance "Current Amplifier"