Unlikely to happen anytime soon. I will have to figure out to manage them myself.
dave
Certainly they can. You cannot get 2 output devices to be perfectly matched.
There are devices were 2 transsitors or FETs are made on the same piece of silicon that can get very close, and chip amp chips all made on the sme die also get close, allowing them to run at really close to Class B since their matching is close, so little to no GM droop and one of the reasons a well done one can sound so good.
One of the reasons things like the NAIM 160/250 were so much ahead of the class in the late 70s, early 80s, and likely why the new Pass XA25 with its new 700w output devices work so well, only 2 output devices (you need 2 for PP) and why SE amps have characteristics many will not give up. None (SE) or little (A/B or A PP) averaging of the signal at very low levels. And given how little power is used at typical levels, very important.
dave
Well, been there, done that! I had to move something heavy and bulky by myself before to go upstairs. If you have a large piece of corrugated cardboard; place that under the speaker box (tape in place if needed) then slide it up the stairs laying on it's longest side. I did this by being below my load so I was pushing it up rather than being above it and trying to pull it. (My logic was that it would be less likely to slip and fall away if I were below it). It sure beats trying to actually lift it up and carrying it up the stairs!
Similar.Averaging out the small stuff. Loudspeakers also have the spacing issue if you are using more than one, whether FRs or a multiway, and in electronics the issue is usually at a lower level.
The same is used to benefit when measuring. Multiple measurements to average out the noise. Given that humans can hear well into the noise thou, only suitable for the “gross” measurements, like trequency response.
dave
When parallel output transistors are grossly mismatched the current will flow through path of least resistance. That transistor may get hotter than the others and the ones with less current flow may suffer more crossover distortion. They will have different responses. Note that the models need to have temp dependent behavior in them and not all models have this.
One can simulate this in LTSpice by tweaking the model files for multiple output transistors to emulate mismatched devices. Re-run the simulation to see what the effect on distortion FFT or the frequency response.
I have done this on line arrays by mildly tweaking the TS parameters of each driver and they simulate the resulting freq response.
The point I was trying to make eight he amplifier is that of a single device (like an IXYS 890w 200V MOSFET) with the same dissipation, current capability etc as say 4x standard IRFP240’s exists, it should work and not have any issues of current matching etc.
Oh....’output devices’ seemed more in relation to drivers than signal generators or amplifiers.....sorry.
Yes....many amplifier designs (AB) rely on phase inversion as a principle of design. In the case of tube amplifier, this function is accomplished with a tube and is VERY susceptible to speaker system resistance. DSP can also play havoc on parallel phase......parametric eQs and passive graphic equalization as well.
A well designed passive crossover and a pair of matched Class A mono blocks.....heaven!
No problem, my fault for going off topic
I'm still none the wiser regards my question about how they can produce a "choir of voices" other than if there is something pathologically wrong with the circuit.I've been looking at drivers for an auratone-style project, which you seem to be doing as well, and have settled on the Seas MCA12RC midrange. Right size, pretty flat frequency response, and looks like it'd play nice in a small sealed enclosure. If you're running it fullrange, I'd imagine you wouldn't care so much about the low-ish driver sensitivity.
If you're looking for a "what people are listening to" reference, how about a cheapo bluetooth speaker?
If you're looking for a "what people are listening to" reference, how about a cheapo bluetooth speaker?
Excellent find! The MCA12RC has everything i’m looking for. Thank you. I’m surprised this driver isn’t on the radar of the fullrange crowd here but I can see it’s one shortcoming in the off axis response.....which is actually a plus in my case.
As to the Bluetooth speaker,.....i’ll be using this monitor for precision monitoring and surgical EQ application so it needs to neutral and clear. I’ll also be using it when adding compression which do at a higher volume than other times...usually 90db for short periods......power compression has to be a non issue which most portable devices succumb to because of the low end extension they try and achieve. I’ll be using this driver in a small critically damped enclosure which the SEAS you suggested will work nicely!
And it seems like many full range builders are interested in higher efficiency and extending the bass response through various non-sealed cabinet designs, which this driver is maybe less suited for.
That's along the lines of how I was thinking of using it. If/when you build something, please share! I'm curious to hear how it works for this purpose.
Thanks but the response is a bit ragged for my purposes....that null and 5db bump in the 4.5-5.5khz range would make it very difficult to get a tonal balance without some passive components or DSP.
FWIW....and if you’re familiar with modern DAW operations.....my mix buss template has three instances of an EQ plugin.....each for a set of monitors I use in the studio. The corresponding plug in gets activated when I’m working on each set of monitors. The EQ profile was created manually for each using a condenser mic at the listening position and REW.....the eQ applied, then re measured and adjusted to get as flat a response as possible......it’s a very effective method and eliminates the need for anymore ADDA steps since it’s all done ‘in the box’.
For you DIY guys that already have an interface, getting a DAW for these functions and using the interface outputs for multi way designs would make for an excellent prototyping workflow.
The final design is going to be a passive horizontal three way for the mixer bridge.....think classic ATC style......all paper/silk drivers in a sealed enclosure. The advantage is a front mounted switch which will bypass everything including the crossover but the midrange driver......so i’ll have that Aurantone Mixcube functionality in one box. I’ve even considered a design with a bit more complexity on the bass side.......a front ported system that can be plugged and another switch for a 2nd network for the woofer to optimize the response for sealed vs ported.....maybe as simple as knocking down the high Q bass response peak of a small sealed box on one side and then disabled when ported. Will take some prototyping but might be worth the effort.......i absolutely HATE having to mix with a subwoofer so if I can get to 35hz for the EDM stuff, i’m good. In looking at woofers, the SEAS CA22RNX looks like the perfect candidate for a 1.5cuft box.
It's not a particularly good driver:
SEAS MCA12RC (Mid-range 4", 8 Ohm, 400 Wmax)
Similar driver, better performance:
MONACOR MSH-116/4 (Mid-range 4", 4 Ohm, 120 Wmax)
-and yes, I was aware of these when I suggested the Fountek..
Last edited:
The CSD plots are both low enough to be of little consequence overall: there is at least 15db before stored energy starts to mar the performance for most of the bandwidth in both drivers.
At higher freq.s the Monacore has it's ridges more "narrow" in bandwidth - which tends to sound "averaged-out" (especially with an axial change), plus it's a declining pressure vs. the Seas (which is higher than its average). Look to the sonogram between the two, the Monacor is overall much more "uniform".
The non-linear behavior is poor for the Seas if run full-range. ..though it's not good for the Monacor either, but it's certainly better to a bit lower freq..
That non-linear behavior is the biggest problem: both are mid.s but the Seas has that added problem (even at 90 db) in a very sensitive range for the ear at 1.3 kHz. It scales horribly with input level.
The Seas's axial performance isn't very good either with 15db difference at 9 kHz between 0 and 30 degrees. Not good in a full-range monitoring driver. (..and I wouldn't consider the Monacore good either, but it is better.)
At higher freq.s the Monacore has it's ridges more "narrow" in bandwidth - which tends to sound "averaged-out" (especially with an axial change), plus it's a declining pressure vs. the Seas (which is higher than its average). Look to the sonogram between the two, the Monacor is overall much more "uniform".
The non-linear behavior is poor for the Seas if run full-range. ..though it's not good for the Monacor either, but it's certainly better to a bit lower freq..
That non-linear behavior is the biggest problem: both are mid.s but the Seas has that added problem (even at 90 db) in a very sensitive range for the ear at 1.3 kHz. It scales horribly with input level.
The Seas's axial performance isn't very good either with 15db difference at 9 kHz between 0 and 30 degrees. Not good in a full-range monitoring driver. (..and I wouldn't consider the Monacore good either, but it is better.)
Last edited:
Yes, but the Seas is over 2% at only 90 db at about 250 Hz. ..and yes, it is laughably bad. (..for a 4" driver with an Fs of 73 Hz.)
It's a mid, and it needs that high-pass filter to perform well - and even then it has that issue around 1.3 kHz.
I does make me wonder what the higher-order distortion is like. Maybe it has low levels of 5th? Or, maybe it's as bad as it looks? Don't know.
It's a mid, and it needs that high-pass filter to perform well - and even then it has that issue around 1.3 kHz.
I does make me wonder what the higher-order distortion is like. Maybe it has low levels of 5th? Or, maybe it's as bad as it looks? Don't know.
Last edited: