Welcome! Need support, you got it. Or share your ideas and experiences.
Return to Planar Speaker Asylum
142.136.193.97
Does anybody have the new curves for the 3.7i? My tube amps didn't like the high impedance spike on the MMGs. The curve on older 3.7s looked a lot better than the MMGs, so it looked like I would be good but I don't have any info on the new crossover.
I had talked to Wendel about my issue but he blew me off and said the frequency response was flat on the MMGs. Well it was with my SS amps, but not even close with my tube amps. I'm going to order a new pair of 3.7s but I sure would like to look at the curve.
Follow Ups:
I suspect the impedance curve is flat since the x.7 models reportedly have a series crossover. With series crossovers, the impedance of the drivers are interrelated so rises in the one driver are countered by a drop in the other.
Note: The constancy of the impedance is true only for 1st order series crossovers, but that is what the x.7's are using.
A series crossover does not automaticly generate a flat impedance curve.
http://sound.westhost.com/parallel-series.htm
The 1.7 is said to have a series crossover and its impedance curve is dropping below 2 ohm at higher frequencies.
http://tinyurl.com/phlgp9p
Roger: I wonder about that. I have the 1.7's .I am using Emotiva mono blocks XPA1: Although have potentially 1000 watts they are not noted for functioning at an impedance in the 2 ohm range; they will not work well with some dynamic speakers with varying loads or even Martin Logan's , the technician told me....I play a lot of organ music and choral music..no problems at all with substantial volume...If there is any distortion it is in the upper middle, not in the tweeter. And also the amps can get hot-so you may be right. But some one will have to open the speaker to if the crossovers are in series.
Well, 2 ohm at high frequencies do not use to be a problem. The level at these frequencies use to be rather low. If not, you will not like to stay in the room...
What happens when you insert attenuating resistors in a series crossover?
http://sound.westhost.com/par-ser-f11.png
Will a attenuating in the tweeter section not upset the rest of the speaker?
I do use an attenuator? Am trying to follow you? But thanks.
I believe the reason for that is because the 1.7 "tweeter" doesn't have any low-pass filter employed. The "super-tweeter" transducer is essentially in parallel with the tweeter at high frequencies and thus reduces the impedance.
Since the 3.7i uses the true ribbon tweeter and a "conventional" midrange we could probably assume there is a low-pass filter employed on the midrange and the impedance level at high frequencies does not drop below 3 ohms.
Dave.
It still depends upon the alignment though....even if first-order. Unless it's "textbook" there will still be peaks/dips in the impedance curve.However, if a flat impedance curve is mandatory without altering the crossover alignment, a conjugate network can be added across the input terminals to make it purely resistive.
Edit: FYI....In fact, I look back in my notebook and it's a 10 ohm, 47uF, and 0.6mH....all in series, placed across the input terminals. That will level the MMG impedance to flat across the entire audio band.
Cheers,
Dave.
Edits: 08/05/14
Thanks for pointing out that the impedance curve is flat only for kosher (symmetric) first order alignments. But this is true for parallel xo's as well - got to love those 1st order filters. In any case the deviation of the impedance from flat is small and tends to drop rather than spike like the MMG, so I think the OP is safe with the 3.7i's.
Yes, the impedance swinging for parallel crossover can be flat too.
The test referenced was for a 1.7 speaker. I'm not aware of an impedance sweep test for a 3.7i..or even a 3.7. Have you seen one?
Dave.
We have not seen the schematic of the crossover of the 3.7 either. How does it work with the damping resistors?
Damping resistors? Do you mean an LCR conjugate network?
Dave.
I mean the level setting resistors for mid and tweeter at the back, http://media.tas.zeitpress.com/articles/images/P4012866.JPG
Ah. Those are in series with the transducer(s) and would only serve to raise the system impedance....and attenuate.
"Damping" resistors would be in parallel, not series and would reduce the impedance load. You don't normally see resistors in parallel (by themselves) in passive networks.
Dave.
Thanks neolith, that was very helpful.
What high impedance spike?
Your chart is not labeled but that looks like the MMG curve? The 1K Hz@23 Ohm rise did not make my Cary Six-pacs happy. It was discovered during listening tests and in searching for the cause, the impedance curve looked like it might be the problem. Swapping out the tube amp with a solid state device completely cured the issue.
No measurements were taken to confirm my listening experience but it didn't really mater, the issue (problem with one frequency) was identified and resolved. The 3.7 curves look favorable but I have not seen data for the new series crossover and I wanted to try the Six-pacs again.
Worse case, I'll dump the Carys and put my 400WPC Mac with output transformers on the job. Did you have something to add?
Edits: 08/05/14
I don't know what you mean by your amps "not being happy." If you're expecting no skewing of the response in relation to the impedance curve then you've selected the wrong amps. :) That kind of response irregularity is unavoidable/inherent when using amplifiers with high output impedance.
Most likely the 3.7i's will not have a flat impedance curve either. The only way to know is to test/plot it. It's a fairly simple procedure if you have the proper test equipment.
Cheers,
Dave.
"I don't know what you mean by your amps "not being happy."
Here's the thread: http://db.audioasylum.com/mhtml/m.html?forum=mug&n=170969&highlight=mmg+russ69&r=&search_url=%2Fcgi%2Fsearch.mpl%3Fforum%3DALL%26searchtext%3Djbls%2Findex.php%2Fadmin%2Fcategories.php%2Flogin.php
Ah yes, I think I remember that thread now. :)
My comment/suggestion to Neo also works for the stock MMG crossover too. The impedance peak can easily be leveled with a conjugate network if this is an issue with a particular power amp.
Dave.
Tube amps use to like high impedancies. 23 ohm is not very high, most conventional speakerscuse to have peaks higher than that. Some electrostatic can have very strange impedance curves. Still not very difficult for the right power amp.
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: