Welcome! Need support, you got it. Or share your ideas and experiences.
Return to Planar Speaker Asylum
87.95.51.10
Hi all!
And greetings from Finland 😊
I'm interested on making my Tympani IV-A's tri-amped with DEQX, preferably using tubes for highs and MAYBE also for mids. I have read this forum (and many many others also) and found a few posts regarding tri-amping true ribbon Maggies, but I still would need some advice, just to be sure I don't break anything... So could someone please help me a bit:
1) All Maggies, Tymps preferably, need lots of power & amps, but of course mostly for the basses. But how much would be "enough" for the ribbons only? I usually don't listen at very high volumes, so would a single ended tube amp with just few watts be enough for them? Or would it be better to go for some bit more powerful one, maybe a push-pull tube amp with 10 or 20 watts of output?
2) How about mids then, do you think that 10 - 20w PP amp would be enough for them, without having to be worried of breaking the drivers etc..?
3) Most tube amps seem to be designed for 4 - 8 ohm speakers. Since ribbons are a bit below 2 ohms, I need to match them somehow for the amp. Many websites say that speaker´s impedance matching should be done with some kind of a (often quite expensive!) transformers etc., but is it actually so in this case with the ribbons (or mids) only? Or could I just add 2-3 ohm series resistor and that's it? If resistor is enough, which kind of exactly?
4) There seem to be two ways for protecting the ribbons when connected directly to the power amp; fuses or capacitors. If I have understood correctly, fuse mostly gives protection for too high volume settings (and thus too much amperes from the power amp), and capacitor for potential DC and quick "snaps" generated by power amps when switched on or off. But what should I do in this case, when connecting the drivers to a low powered tube amps; fuse, capacitor, neither or both of them?
Also, all other tips & trick regarding tri-amping & tube amps with Maggies are highly appreciated! :=)
Just to give some point of reference to your question regarding running T-IVA's with tube amps. I have rebuilt a set of T-IV (non-'a') about a year ago. I'm running them with the original XO boxes, and have modified the internal crossovers for a flatter response and to 'fix' the weak midrange that was the main criticism of them and why Magnepan updated them to the 'a' revision. I have not re-capped the crossovers and am running them full-range, not bi-amped or active XO.
These speakers sound fantastic without seeming to be 'power hungry'. I've run them with Class A/B (push-pull) tube amps that are only capable of about 60WPC max, and likely lower power with the lower power tubes I am running (Svetlana 6550's vs. spec'd KT-120's).
My system is not in a huge room, so I don't need a large amount of power to drive them to levels that are 'uncomfortable' in volume. I have run the various tube amps up to levels where they begin to clip, but tube amps clip 'softly' and they have not caused any problems with the ribbons. By the time the amp is running at that volume level, my neighbors would be getting upset and my ears wouldn't be able to take the in-room volume for very long either.
When I first started on the rebuild of these, I had blown out one the tweeters (which were old/loose in their frames to begin with) because of a tube pre-amp microphonic feedback loop that overpowered one of them. The tweeters were not separately fused from the mids (Magnepan's original design). If your tweeters are not separately fused, make sure you order the update kit from Magnepan for them.
I don't think you'll have trouble driving the ribbons/mids with tube amps that are 20-30W/amp section. If you can drive the entire set with amps that are 100WPC or less at comfortable listening levels, you should be fine. Keep in mind that getting rid of passive components should also increase the 'efficiency' of each driver as well.
Thank you very much for your comments UncleMeat! :=) It's quite interesting to see how different opinions and experiences people have on this topic; some feel that there should be hundreds of watts to get best out of IV's, while some, like you, are happy with way below 100W.
But maybe this just points out that it's always subjective, and depends highly on the room size, preferred music style, wanted volume levels, maybe also person's hearing, etc.
Since my room is not very big and I like to listen on a near field with quite reasonable volumes, I think I will give tubes a try on mids and/or highs. If they won't work, it's quite easy to get rid of high quality tube amps anyway.
Happy new year for all you Magne-heads! :-)
Yes, and another way to go about tri-amping would be to split the bass panels and amp them separately, but leave the mid/high pair together with the high-pass amp driving them via the internal crossover. That would probably give the most even load distribution among the 3 amps.
On my T-IV the two bass panels are different. One is a 'low' panel, and the second one is a 'mid-bass' panel with tuned sections for the higher low frequencies. I'm not exactly sure how high the mid panel goes but it -could- be semi-full range (up to maybe 5kHz?) if a frequency sweep was done to determine it's properties. Putting my ear next to it while playing music tells me that there is -some- higher frequency sound coming out of it, but probably largely blocked out by the low pass crossover.My thought/potential plan was to separate the two drivers by disconnecting the wires going to the mid-bass panel from the low-bass panel. That gives the option for powering them individually, which would set up a tri-amp if you kept the T-IV midrange/tweeter paired, or quad-amp if you split the mid/tweeter and amped them separately as well.
Edits: 01/13/25
How can they be different when its a 3 way system?
Technically T-IV's are 4-way speakers, reduced by a 3-way crossover. The high-pass goes to a second 2-way crossover for the ribbon/tweeter, while the low-pass goes to two bass drivers on separate panels. Those two bass drivers are 'tuned' to respond to the various frequencies in the low-pass signal, one does everything below 100Hz (or so) and the second one has 3 tuned sections that cover above 100Hz to ~5kHz.Those bass drivers could be driven separately on single amps, doing a freq sweep would tell at what frequencies to set the crossover points/slopes at.
Edits: 01/14/25
Here a comparsion of the two bass drivers. The one the left normally sits in the center of the speaker, it is lower tuned. The one on the right is higher tuned. Still, these are bass drivers not very high performing at higher frequencies even if their response is wide band. The basses of the older T-IIIA is flat up to about 4 kHz but does not sound like that.
![]()
Edits: 01/15/25
Yes, understood Roger. My T-IV's don't have a tuning button on the 'low' panel. The mid-bass panel looks similar to what you've posted but the center button is higher, closer to the top support bar than the second one down.
T-IV is different, thinner magnets too. The low bass driver tuned to 43 Hz and the mid bass to 57-75-88 Hz (from bottom to top).
![]()
Edits: 01/16/25
That sounds like bi-amp to me, have IV,a as well
Happy New Year!
Use an active crossover and it gave me an additional 1.7 dB in the bass from my 3.6. T-IV and T-IVa have similar passive crossovers. This will lower the power for the drive by 33%. There is less to gain from the midrange, maybe less than a dB.
I actively tri-amped my IIIAs (and their replacement ... my 'Frankenpans' - which used T-IVA mid panels and ribbons) for 25 years without using a series cap on the ribbons. I didn't blow a ribbon, ever - although it should be noted I was using ss amps which had a lot more power than you are thinking of! (120w into 4 ohms - and built to be able to drive 2 ohms!)
I think driving even 3 ohm ribbons with a tube amp would be non-optimal; perhaps you should use 'autoformers'?
Message me if you are interested. Thanks.
Great choice, and super easy to do. You do have to pad them down unless you triamp.
in this thread on ak:
https://audiokarma.org/forums/index.php?threads/ehooves-current-rigs-updated-9-9-24.694546/
Tells what the tube amps are for the mids and highs.
I would say that tri-amping is a bit overkill, there is really not much to gain. Bi-amping is a must and so is DSP for correcting the frequency response. The passive crossover mid-tweeter may be given different slopes, moving the crossover point a bit higher. The 2 Ohm ribbon often need to be reduced in level, at least 1 Ohm or even more in series with it. Then you have 3 Ohm or more up there. Have you checed that your mids are in good shape, that the double-sided tape is holding teh Mylar? I prefer to have foil conductors on the mids. It will lower the output a bit and therefore you need to reduce the level of the tweeters (and raise their impedance...) even further. I would recommend at least 100 W/4 Ohm for the mid/tweeter section.
But since I already have DEQX processor with 3-way digital crossovers, I really would like to get rid of passive crossover components on the signal path. With DEQX I can also easily test different crossover points and slopes, to see (or hear ;=) which setup sounds the best? Different gains/levels can also be easily matched, frequency response & phases corrected etc.
And my mids are be fine, thanks :-)
Yes, I know but it is a bit risky to run the ribbon without a capacitor in series. I also have the possibility to run up to 4-way but do not.
Well, that problem could be solved by adding a capacitor in series with ribbon, wouldn't it..? :-) This is exactly what I'm now trying to understand; is it really "too risky" to connect ribbon (and/or mid driver) directly to the tube amp, or would it be a good idea to use some good quality capacitor as a protection?
Blowing a ribbon, compared to a "normal" dynamic tweeter, shouldn't be a catastrophic situation anyway, because it's quite easy to fix with manufacturer's repair kit. But of course, if it is clear that it WILL blow up sooner or later, it might be wiser to prevent it from the beginning.
I wouldn't run the ribbon without a capacitor. It isn't just a matter of DC. It's all too easy to make a mistake, and then your ribbons are fried. As Roger says, you can compensate for the cap electronically if you wish. But I agree with him that since you want a cap, you might as well leave the existing crossover in place. It's easier, cheaper, and more practical.
I do run my system three-way, but that's because I've replaced the mids with Neo 8s, which are much more efficient than the ribbon and bass panels. I had them padded down for a while, but it was fairly ridiculous.
By the way, I second Roger's concern about the adhesive on the mids. Mine kept coming loose, and I had to press them back down. Don't know how he solved that problem. I agree too with his suggestion to replace them with foil if you can.
Finally, I don't think tubes are the best way to drive Maggies. They're just too power hungry. Maggies require several hundred watts to "sing." (Of course, it depends on how loud you like to listen -- I like to turn them up occasionally and listen at realistic levels, but most of us don't do that most of the time.) I had a Parasound A-21 on my bass panels for quite a while, and now I have a capable Class D amp and the Parasound is driving the tweeters -- overkill, I know, but I already have the Parasound, along with an AHB-2 on the mids -- not a large amp, but it's fine for the highly efficient Neo line.
Thanks Josh!
I know that Maggies ARE power hungry. But I have always thought that it's mainly the bass drivers which needs the power, not so much mids or highs, but does anyone have any facts/experiences on this?
The whole speaker is inefficient. The IVA manual claims that 200 watts will driven them to 110 dB SPL at the listener's seat.
Bass, as you suggest, typically requires the most power, followed by midrange (though it isn't that much lower), with the tweeter needing much less. I'd probably split the difference and go with equal power for the bass and mid/tweet combo.
My personal preference is for lots of headroom, IMO, a solid-state amp should never clip -- the limit should be set by the speakers and their fuses and your ears.
Yes, the mids/tweeters do need a fair bit of power and the impedance is rather low. I have seen them driven by Krells on the basses and Audio Research on the rest, not sure what model, maybe D-90. If the tweeters are padded down (with a series resistor) their impedance will not be a problem for most tube amps. Padding down will be a must if you go for foil on the mids (it lowers the output on them).
My mids are still not fixed... I will need a replacement of the Mylar as well.
Why not go with the Neo 8 clones? They slot right in to the existing baffle.
Which clones exactly do you mean? I thought that NEO8's or any similar drivers are not available anymore...?
The clones are apparently pretty good, they've been compared side-by-side. Emailtim here has some.
In one way they're better -- their impedance is 8 ohms so you can make the line come out to 4 ohms. Not that it's super important given how efficient they are.
The Tympani IVA can accommodate only 7 of them in the original baffle, but they are the same width as the opening so really easy to mount.
I like the original IVA midrange, but these are head and shoulders better, at least the original mids are.
Maybe these https://www.parts-express.com/GRS-PT6825-8-8-Planar-Mid-Tweeter-8-Ohm-272-126?quantity=1The GRS seems to have a bump at the fundamental resonance, the BG had no such bump.
As you will need 14-16 of these, it will not be a cheap option for us in Europe.
The original BG Neo 8 were better but became worse in its last years of production. The European distributor droped them because of large variations between samples.
Edits: 12/04/24
Thanks Roger! 16 pcs of those would not be "cheap", but I'm afraid that if I have to send my mids from Europe to USA (and back) for refurbishing process, it might cost even bit more...?
Small bumps wouldn't be a problem anyway due DEQX, if the sound quality over all would be better than stock. But does anyone here have any experience on those GRS's, how would line of those compare to the TIVa's original mids (or NEO8's)?
A line of the original BG Neo 8 have some users, Josh is one of them. Refurbishing Magnepan can be done in Europe (some guys in Germany). Sending them to Magnepan will be very expensive...
Edits: 12/05/24
Check your messages please....
Actually, now I found one nice tube amp DIY kit, which gives around 28 watts of output AND can be easily modified for 2 ohm outputs also, so low impedance shouldn't be a problem at all! :-)
Also, drivers' different sensitivities/volumes is not a problem either, since I can easily match them on the DEQX.
But Roger, when you say that "mids/tweeters do need a fair bit of power", what exactly do you mean, how much is "fair" for highs, how much for mids? Have you tried some lower power (tube) amps on them, what happened, how was the sound? I don't need "realistic" volumes anyway, I usually like to listen at quite mediocre volume levels.
And I understand well that it might be hard to say any exact wattages, "enough power" depends on so many variables. But if someone here has already tried any low power amps on 3-way setup, I'd really would appreciate to hear some real life experiences, both good and bad.
I have used a Denon minireceiver for my 3.6 for a while. Output was like 2x45 W in 4 Ohm. Later, I went to something like 2x130 W and then
4x260 W. The peak power for higher frequencies is almost as high as for the bass. Most Magnepans are not as efficient as the numbers on the Magnepan's spec. I have measured something like 82-83 dB/2.83 V.
Roger, are you sure about this: "The peak power for higher frequencies is almost as high as for the bass"??
For example, at PS Audio's web page it is said that "It takes a ton of energy to reproduce bass notes, but not a great deal as the scale goes upwards. Fact is, most tweeters never see more than a few watts."
https://www.psaudio.com/blogs/pauls-posts/tweeter-power?srsltid=AfmBOoolGcWuHJHrs-ksa08d11QtZEmouUrLOXbaKXvLxV0gsN5o31C1
Here are two examples:Michael Jackson - Don't stop til you get enough, the upper curve
![]()
Michael Ruff - Wishing Well, orange below 100 Hz, blue above 100 Hz
![]()
Edits: 12/03/24
Thanks again Roger, but I don't have a clue what these images should tell me regarding the tweeter's power needs? :-)
" My advice is still to have similar power on basses and mid/tweeter " ... I would agree.
I used (ss amps):
* 180w into 4 ohms on the bass panels
* and 120w into 4 ohms on the mids & ribbons (3ohm ribbons).
The one on top, shows the same level from bass to tweeter.
The second one, shows a higher power demand for frequencies above 100 than below, 3 times more.
These are peak level not average. Musik is peak level, not sinus.
My advice is still to have similar power on basses and mid/tweeter.
Very delicate and tedious work. Of course, entire replacement units can be had, at a cost.
Mark in NC
"The thought that life could be better is woven indelibly into our hearts and our brains" -Paul Simon
Adding a capcitor in series with th ribbon is almost a crossover. Why make things more complicated with adding another power amp? The cost of a dual channel power amp is higher than using a passive crossover.
Good point about the extra cost of another power amp. My main point for tri-amping was the DEQX, which easily allows trying & listening different crossover points & slopes (for example even 96db) for the best sound. DEQX will also allow frequency- & phase corrections for all drivers, and for the best outcome, every driver should be measured separately on the near field.
Minimizing the number of passive components from the signal path is of course a plus also.
FWIW, I quad amp and use both fuses and a cap to protect the ribbons.
I do NOT use tube amps so I can't speak to their unique properties/considerations.When you design your ribbon XO, you will have to take into account both the slope and phase rotation caused by the cap you select in addition to your DEQX XO, should you chose to use one.
I designed a FIR ribbon XO that includes a filter which electronically erases the slope and phase rotation of the cap, so when playing, the cap is essentially out of the circuit leaving just the digital XO. This guarantees the Mid and Ribbon XO's ideally sum without the cap's signature tainting the XO region.
.
2022/03/30 Historical Records CENSORED
Edits: 11/18/24
This post is made possible by the generous support of people like you and our sponsors: