I heard this speaker with a TDA2003 with basic current feedback and it still gets better sound although it has a Faraday ring which minimizes already distortion.
https://www.diyaudio.com/community/...loudspeaker-sandwich-cone.402917/post-7443470
But I tested first time simple resistor method with 4 ohms and 1.5 microfarad in parallel bringing very good sonic result too with this 8cm fullrange loudspeaker. So for this speaker I bought an efficient D amp and prepared some resistors for a 2.1 system.
https://www.diyaudio.com/community/...loudspeaker-sandwich-cone.402917/post-7443470
But I tested first time simple resistor method with 4 ohms and 1.5 microfarad in parallel bringing very good sonic result too with this 8cm fullrange loudspeaker. So for this speaker I bought an efficient D amp and prepared some resistors for a 2.1 system.
Yes. But your focus seems to be on the "Q" of the bass bin, rather than the balance of the audible spectrum. This is a mistake, I believe, as I believe you're missing the real advantage of current drive. What is the real advantage of current drive (transconductance) amplifiers? Insensitivity to driver-induced back-EMF (i.e., flux modulation, solenoid forces, back chamber-induced microphonics, small room horn-amplified reflection microphonics, etc.).
More discussion here from this forum from 20 years ago, including Mr. Pass (and his First Watt F1 and F1J transconductance amps): Transconductance power amplifier for current-drive loudspeaker?
What kind of drivers benefit the most from being driven by transconductance amplifiers? Drivers that have a very wide operating bandwidth, and as such are much more susceptible to modulation distortion of the FM type. One example is full-range lightweight-diaphragm drivers (e.g., Fostex, etc.), which I personally have little use for due to their extreme limitations in amplitude output, especially bass frequencies.
My personal preference is high-efficiency drivers with very wide bandwidths. In general, these are larger throat diameter/full-range compression drivers on large straight-sided (pyramidal) horns. One more recent example is the Celestion Axi2050 on a 1-m wide 90 x 60 degree straight-sided horn.
Chris
YES >
and that would be a driver performing on its own merit or lack there of.
To my understanding, there seems to be a lot misunderstanding regarding this subject. (what causes what)
Feedback loops, and associated 'sensors' are used to correct something that is distorted in the first place,
and as we know, excessive negative feedback causes other distortions > such as Transient Intermodulation Distortion.
If feedback was truly the panacea for good sound, then we would see much more of it > like motional feedback,
however this is not the case.
It sort of reminds me of the saying > "Putting lipstick on a pig".
If a driver can't handle thermal increase, then that is a limitation of the driver!
Could you please elaborate, especially since the misunderstanding is a lot ?
Feedback loops are primarily meant to exercise control over state variables and their effect depends on the type of control used. Distortion correction (reduction) is just a secondary consequence of that control.
As already mentioned in #9, if the heat generated is removed (using vents or whatever), there should be no further increase in the VC temperature. But when it is not, the 'thermal increase' leads to more 'thermal increase' and so on until the thermal limits of the driver are reached. This is a general tendency.
I think we should let the OP explore the loudspeaker possibilities for current drive, like flat impedance etc. What I meant was that we just pretend for the time being, that a 'true current drive amplifer' already exists, so that the OP is able to continue with the discussion that pertains to the loudspeaker.
Last edited:
OK but we need to look at some fundamentals.. We are not really interested in Current Drive or Voltage Drive.
What we want is the Soundfield near your ears to be a good approximation of an Electrical signal from your CD player etc. Current or Voltage drive, speakers bla bla are just some of the intermediate stages which might facilitate this.
Rice & Kellogg, inventors of the modern moving coil speaker in 1925, chose to solve the problem with Voltage Drive and 'heavy' mass controlled moving coil speakers where the moving mass completely dominates the air load. In doing so, they invented Bass Response (ie an extended flat frequency response) Before them, all efforts were just to get sound louder, not better. ALL modern speakers are designed to this paradigm. Da Thiele-Small pseudo gurus would do well to read their 1925 paper for insights into what makes a good speaker.
https://www.aes-media.org/historical/html/recording.technology.history/rice-kellogg.html
If you want to do the same thing, with Current Drive, you need the moving mass to be dominated by the air load. This is difficult with moving coil units but may simply be cos very few people have worked on it.
It's 'easier' to do this with an electrostatic speaker but all current ones have been fudged to work with Voltage Drive. Walker hints at how to do this properly in https://www.aes.org/e-lib/browse.cfm?elib=3947
In my book, 'flattening the impedance to get flat response' bla bla is just fudging. Today, you might as well use DSP to flatten the response. But fudging may be what yus guys want rather than addressing 'loudspeaker especially for current driven amplifiers'.
I've built and measured several 'Current Drive' amps in da previous Millenium, long before Merillainen et al and this century have done loadsa LTspice simulations. None have had Zo > 100R at 20kHz AND been stable except possibly Mills & Hawksford.
Since you frequently bring up the stability aspect, I guess you're talking about linear amplifiers. The main issue with linear systems is that there's only so much loop gain you could put on before hitting instability. If you're not satisfied with a linear system, you could always try the non-linear control techniques like hysteresis, sliding mode control etc. to obtain better control performance.
I would let the OP reply to this one.
A quick look at a simplified schematic of an amplifier with current drive usually shows that the speaker is put into the negative feedback loop. Normally, the loop is meant to be small to reduce RFi and so on. But with current drive it could get very messy.
A better bet could be to configure the output devices as open drain, and simply live with the highish distortion.
Looking at a single-ended class-A topology as a base, maybe modulate the constant current source, a la Pass' 'Aleph'? Cascoded JFETs? A compound FET in open drain? High voltage FETs with low current and a step-down transformer?
A better bet could be to configure the output devices as open drain, and simply live with the highish distortion.
Looking at a single-ended class-A topology as a base, maybe modulate the constant current source, a la Pass' 'Aleph'? Cascoded JFETs? A compound FET in open drain? High voltage FETs with low current and a step-down transformer?
Yes, and either way it is just bad/wrong terminology.
A battery charger is usually just a rather simple 4/6 diode rectifier anyway.
And has nothing to do with the subject at hand really.
Not necessarily, because it is the 'loop area' (not the loop length) that needs to be small to reduce EMI / RFI. If the speaker lead and its return were to be twisted together, this loop area (A) and therefore the flux (φ, due to the magnetic field B), could be reduced. A shielded cable maybe used for simplicity..
Also, the expression for the induced emf is V = BAω = φω, where ω is the angular frequency.
Hope that helps.
I was referring to the more sohpisticated current-controlled battery charger that delivers a metered current to a battery say, 1A to a 10Ah battery (0.1C). As the battery picks up voltage, the charger's voltage is continuously adjusted to maintain that current (1A) until the 'charged' condition is reached e.g. 14.4V for a 12V acid battery.
The current-controlled amplifier rejects the effects of the circuit impedance (Re, Le etc.) and the back-emf in just about the same way the charger's rejects the battery voltage and internal resistance. The charger and the battery voltage are thus analogous to the amplifier and the loudspeaker back-emf respectively, with the only difference being DC vs AC operation.
I refrained only because this thread was meant to discuss loudspeakers as opposed to amplifiers.
Hi kgrlee,
Kindest regards,
M
Do you think that this is the reason that the AMT drivers, e.g., Heil have a flat impedance?
Kindest regards,
M
As I am a fan of fullrange loudspeakers I tested amps converted to basic current feedback.
This type of loudspeakers profit a lot from this amplifier topology.
One possible answer concerning ideal choice for a current driven amp is using quality fullrange drivers.
I my opinion fullrange drivers have been vastly underestimated and are constructed in the mass production on a very low level:
I struggled a lot to overcome the deficiencies of fullrange loudspeakers by making stiffer and less resonant cones than "only paper" drivers.
https://www.diyaudio.com/community/...loudspeaker-sandwich-cone.402917/post-7468479
Learning here from others about current driven amps and how to build them really helped me a lot to make my designs sound better than I expected.
This type of loudspeakers profit a lot from this amplifier topology.
One possible answer concerning ideal choice for a current driven amp is using quality fullrange drivers.
I my opinion fullrange drivers have been vastly underestimated and are constructed in the mass production on a very low level:
- usually only little excursion capabilities
- too lightweight cones having too much resonances due to a lack of stability
- very rarely faraday rings helping getting distortion down and helping getting more high frequencies
- no use of current drive amplifiers which are ideally suited to compensate for lack of bass and highs in a natural way
I struggled a lot to overcome the deficiencies of fullrange loudspeakers by making stiffer and less resonant cones than "only paper" drivers.
https://www.diyaudio.com/community/...loudspeaker-sandwich-cone.402917/post-7468479
Learning here from others about current driven amps and how to build them really helped me a lot to make my designs sound better than I expected.
@Freedom666 Any thoughts on spider suspension? One idea that comes to mind is to replace the usual spider with a ring with 3 eyelets and strings under tension.
If you want your amp to sound different, making it non-linear (ie distort) and have iffy stability will certainly do that. But such amps do poorly in DBLTs 🙂
Have you got links to schematics of some of your amp designs?
I've only done one listening test on a Heil in da previous Millenium. We weren't impressed so didn't measure it. Have you got an impedance curve of the AMT itself? ie not the complete speaker.
@abstract
From loudspeaker history it's known that Ted Jordan had for his first 10cm fullrange full metal cones an alternative suspension which was said it had better linearity than the standard yellow spiders.
From Voicepoint (Pfleid patent loudspeakers) I heard that simply yellow spiders from 25cm drivers were taken for 20cm drivers to get more linearity.
Some 20cm high end fullrange drivers, forgot the name, they made more lightweight yellow spiders with a more open woven mesh having less losses than thicker material.
In my eyes very important are mellow surrounds and spiders and stiff cones for good impulse response. Middle stiffness of suspension is still ok.
From loudspeaker history it's known that Ted Jordan had for his first 10cm fullrange full metal cones an alternative suspension which was said it had better linearity than the standard yellow spiders.
From Voicepoint (Pfleid patent loudspeakers) I heard that simply yellow spiders from 25cm drivers were taken for 20cm drivers to get more linearity.
Some 20cm high end fullrange drivers, forgot the name, they made more lightweight yellow spiders with a more open woven mesh having less losses than thicker material.
In my eyes very important are mellow surrounds and spiders and stiff cones for good impulse response. Middle stiffness of suspension is still ok.
Hi kgrlee,
But my question is, whether you think that it is caused by the air load dominating the moving mass.
Kindest regards,
M
I did, but my laptop battery died, so I lost it. However, attached is perhaps a more reliable measurement.
But my question is, whether you think that it is caused by the air load dominating the moving mass.
Kindest regards,
M
Thanks for this. The impedance curve shows the mechanical damping, which may be mainly acoustic (ie the air load which is what we want) dominates the electromagnetic damping. The shape also suggests electrodynamic damping is very low .. especially in view of the HUGE magnet structure.
Du.uuh! I've just seen their Qms, Qes & Qts numbers which say exactly this.
It also show some resonances / leaks / cavities in the structure which give the double peak and also some of the smaller kinks.
Otherwise suitable for Current Drive. The double peaks only cause narrow 2dB peaks in response.
Last edited: