Welcome! Need support, you got it. Or share your ideas and experiences.
Return to Planar Speaker Asylum
203.12.195.242
In Reply to: RE: wattage, more than low impedance.. Tweeters use a few watts max. posted by 3+4=5 on February 15, 2017 at 12:34:49
True - and up until last weekend, I would've agreed with you.
But I found something out about feeding HFs into a 2 ohms load that has changed my mind.
I run triamped Maggies - so I have an 70w (into 8ohms) ss amp driving my ribbons directly. The amp is stable into 2 ohms - ie. it hasn't blown up! :-)) - and my treble sound is great (or so I've always thought!!).
But I recently did some sweeps using REW and a calibrated mic ... and noticed that the FR graph showed a steep drop-off after 16Khz. (Given the mic is calibrated, the FR should be flat to 20Khz.)
So I did some crude measurements, driving my tweeter amp directly (bypassing the miniDSP XO) with a signal generator and measuring the voltage at the ribbon terminals.
To my surprise, the voltage I read at 16Khz was much lower than the voltage at 6Khz ... and at 30Khz, it was down a lot more! This says to me that my tweeter amp - even though it is "stable" into 2 ohms - cannot perform at HFs into a 2 ohm load! :-(( I have been told by an amp designer (not the designer of my amp!) that this is typical of the 'power envelope' of Mosfet power amps.
So I am going to borrow a transistor-based power amp which is renowned for being able to drive 2ohm loads and see if the REW sweep becomes flat to 20Khz!
So my point is - just because an amp:
* has bags of power for a 2ohm driver
* and is "stable" into this load
... doesn't necessarily mean it will perform well into that 2ohm load.
Andy
Andy
Follow Ups:
Any idea of Phase Data?
Too much is never enough
.
You know... I just stumbled across this because my pre went down then I started to question the synergy of everything. The psaudio directstream going straight to the amp sounds pretty good. My power supply has been shut down for weeks on my Cary. I see something where the AC connects which indicates a fuse behind some plastic piece which I have to figure how to pry open. In any case, if it is a fuse - I'll figure it out and report back.
Very interesting, I would not have expected that. Did you try measuring the current through the tweeter as well? I wonder if you added a 1 ohm resistor in series would you see the same voltage drop? By doing so you could also measure the voltage across the resistor and calculate the current.
OTOH you are probably stone deaf above 16 kHz :).
I married the perfect woman. The downside is everything that goes wrong is my fault.
That may well be true, neo, but let me relate a story which indicates to me, anyway, that we can respond to much higher frequency sounds than 20Khz.
The first version of my single-ended, jfet-based 'Muse' phono stage that I built about 5 years ago was regarded by the people who listened to it (comparing it to other phono stages) as being slightly rolled off in the treble. This first version was a circuit that came out of Silicon Chip in the late 80s.
A mate modelled the circuit in LTspice and told me that the simulation showed a LP roll-off at 120Khz . This was caused by a cap which had been put there for stability reasons ... but probably wasn't necessary.
So I removed this cap - and the phono stage no longer sounded dull in the treble! So we could hear the effect of a LP filter at 120khz! who woulds think it!! ;-))
Re. current through the ribbon - surely "V = IR" where:
* R = 2, and
* V = 2.3v at 6khz
... means I = 1.15a?
And W = I^2R means the amp was delivering 2.6 watts at 6Khz.
Whereas with V = 0.9v at 30Khz, I = 0.45a and W = 0.4 watts.
Andy
I had that sort of issue but didn't measure the voltages with the Bryston 4B NRB which I tried on the tweeters. It is rated to 4 ohms and has oodles of power, but only "stable" to 2 ohms. Though it did drive them, it ran hot and sounded thin. When I had the option I swapped with a MOSFET Fosgate amp that weighs the same but has 1/4 the power rating. Also rated to 4 ohms and manufacturer suggests cooling fans when running into a 2 ohm load. It was run with and without a 1 ohm resistor in series. Without a resistor the amp got hot and images became tiny. When it got hot to the touch I would turn it off and put the resistor back, which is how I ran it most of the time. But for that one time when I forgot and ended up burning the amp. I have another which I use on occasion, and always with the extra resistor, which I also use with my tube amp. The only amp that drove the tweeter with no problem at all was the Classe DR9. Which is rated to 2 ohms. Ran a bit warm just as it did on the 6 ohm mids. The PSE Studio IV didn't stress on the tweeters but ran hot.
I did hook up the Crown just to see what would happen. Which was nothing in particular, it sounded like the Bryston on top, which is as expected. But there was no heat, the fans never turned on and the load light was never lit. But then it is rated to 2 ohms and stable to 1 ohm.
So I see that a low impedance load can be a problem for an amp not designed for it regardless of power rating. Experimenting with different resistors on the Fosgate amps it was rather obvious that it did best when the load was over 4 ohms. Even though it was losing nearly half its output to the resistor it was never stressed and never ran hot playing single ended (40W) and definitely no problem in bridged differential mode (80W), which did not sound as good as it did in single ended mode.
BTW if you are wondering why I didn't use the resistor all the time it was because it was the OEM Maggie ceramic resistor and it sounded like $^#$T. It took me time to admit to the fact and order a nice mills wirewound piece. Once I had that at hand I did not run the tweeter amps without the resistor, except for the DR9 which didn't need them.
For the OP
Thinking further then it seems if the high freq impedance drop is a concern for your amp''s high freq performance then you should find the 20.7 a substantial improvement. But then, I don't know that it actually is an issue for it. I second Neo's suggestion to take the amp with you to the dealer's to try it out on the 20.7. While there ask him about any "sanctioned" conversion of the XO to biamp.
Though there are many other factors involved, I think that if your amp sounds brighter on the maggies than it sounds on the MLs then it is likely that its upper end FR is drooping into the low impedance load of the MLs.
I discussed this in detail with Wendell when I got my 3.7i. Basically he said "why take a step backward when we already optimized the crossover". In other words, NO.
I married the perfect woman. The downside is everything that goes wrong is my fault.
Agree with what Satie said. Now, some would say Magnepan could've chosen to continue on with the parallel XOs of the 3.6 - just changed the slopes - but they didn't ... so the series XO version must sound better.
The truth of that I cannot comment on - but I'd stake my 'left one' that a 3.7 that had its passive XO excised and replaced by a 3-way active XO ... would sound much better! :-))
Andy
PS: and if you ask Wendill about replacing a Maggie's MDF frame with a hardwood frame, he will give you a similar brush-off comment. :-))
No disagreement there. I am just not up to it at this time.
I married the perfect woman. The downside is everything that goes wrong is my fault.
We are talking about copying the XO electrical performance into a parallel line level XO, not substituting a new design. Which is hopefully what others who biamp are doing.
Exactly the perfect place to Start.
Plenty of time ot clown around later!
Too much is never enough
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: