Help me choose which of these amps for ribbon tweeter

[NiToNi]

I am putting together an open baffle project with the LCY130 as the tweeter:

LCY-130 @ Madisound

Most of the budget has already been spent on the mid & bass section (Supravox 215 S bicone with PrimaLuna PL 2 and AE OB15 with ICEpower 500ASP) so I'm looking for a cheap amp to drive the ribbon, which I expect will be asked to do its job from 5kHz and upwards (depending on experimentation).

Ideally I would have wanted to use an OTL tube amp for this duty but even the cheapest ones are beyond my reach for this project. Hence, I was turning my attention to one of the cheaper chip amps on the Bay. I have narrowed it down to one of these, of which one is actually a regular SS amp and another a T-amp.

LM4562 + 2SA1943/2SC5200

Original LM3886, NOVER 6800U/50V golden Caps, 35A rectifier

LM833 + LM3886

NE5532 + LM3886

Tube 6N11 + LM3886

Tube 6N3 SRPP Preamp TDA7294

TA2022, ALPS, ERO+Siemens+CDE+Panny FM caps, KOA resistors, OMRON relays, ONSEMI rectifiers

TA2022, UPC1237, NE5532, ALPS

I have no experience whatsoever with these designs so I would appreciate if some of the more experience forum members could guide me in the right direction given my specific application and the (rather limited) info given in the various item descriptions. Price-wise between these, I'm indifferent. Power-wise they are all ball park where I want/need to be. My gut feeling tells me I should go with either of the two first ones in the list...but what do I know

Needless to say, with a ribbon I am looking for the amp with the most distorsion and grain-free, transient-fast and detailed, extended (beyond 20kHz) and airy character. Or is that too much to ask from any in this line-up?

[Nuuk]

Well you are on the right road with the OB's, and have chosen some good drivers. But your choice of amps is going to spoil it all!

You cannot mix class D/T amps with class A/AB. So, I would forget the ICEPower amp and use an LM3886 for bass (I use one with a Hawthorne Augie 15 inch woofer). An LM3886 should do the for the tweeter too.

A s/h valve amp may also come into your budget.

Also be aware that what you buy off Ebay may not be all it seems (fake/recycled/cheap chips and components)

And try and stick to a single valve pre/buffer for all the amps, rather than adding 'different flavours' to each amp/driver.

[Pallas]

Originally Posted by Nuuk

You cannot mix class D/T amps with class A/AB. So, I would forget the ICEPower amp and use an LM3886 for bass (I use one with a Hawthorne Augie 15 inch woofer). An LM3886 should do the for the tweeter too.

That's silly.

[NiToNi]

Thanks, Nuuk!

Hmmm… I would’ve thought it was indeed OK to mix different amp technologies in an active system, perhaps not even a bad idea since it would allow me to benefit from the strength of each (class D for bass, valve for midrange etc). After all, I’ll be mixing different type of drivers here, which ought to be more audible if anything, no?

Frankly there is also a budgetary reason for my choice of different amps as I happen to have the PrimaLuna and the ICEpower modules lying around and would like to recycle them in order to keep the overall project cost down. Besides, I like the way the PrimaLuna sings

Not to blur my own thread topic into one of OB design, but it may help you to know that the OB will have two AE OB15 per side, with the one closest to the floor low-passed at 80-90Hz and the other at 120-140Hz (i.e. a 3.5-way). High-pass at Fs. One ICEpower 500ASP per woofer.

I’m not sure an LM3886 based amp is powerful enough to drive the woofers to their Xmax at Fs, something I want to do so I can equalise a flat response over their passband (30-120 Hz) while maintaining acceptable SPL. The OB15 in an infinite baffle is not reaching its Xmax (18.5 mm) at Fs (29 Hz) until input power of 400W @ 114dB. Given the ICEpower modules output ~200W into an 8 ohm load and my baffle is down 12 dB from 120 Hz at Fs, the two woofers will output 105 dB @ 30 Hz. Then if I sit 8 feet away, I’ll get 98dB net of dispersion/distance loss. All without taking room boundaries into account of course, but still – I don’t want to give up any more acoustic output considering I’d like a bit of a fuller-sounding house curve with the woofers about 3-5 dB higher in level than the full-range driver.

The PrimaLuna is also a good power match with the Supravox.

The efficiency of the ribbon tweeters is rather low in real life (~87 db) so a bit of juice is needed to ensure they keep up with the rest. A valve amp would be great but I guess it would be difficult to find one even second-hand for the same kind of money as those eBay amps I am looking at.

There won’t be a preamp per se. Each of the eight amps will feed off an RME sound card housed in a PC convolving Audiolense XO filters (and equalisation). Well, the midrange channels will actually go (as S/PDIF) via an MHDT Havana to the PrimaLuna. Also, digital source into the PCXO and digital attenuation of volume (24 bits) from reference level and down.

Both the PrimaLuna and the eBay tweeter amps have volume pots so I could match levels prior to building the digital XO filters and thereby maximize digital resolution. If it turns out that it is the woofers that require padding (imagine that, after 12db attenuation) relative to the midrange & treble, I have a Prometheus TVC lying around too that perhaps could do the job.

Here are two other designs that seem of higher build quality although at double the price:

Valab AM3: Three LM3886 in parallel

Marantz MA-9S2 clone

I have no idea whether they are worth it in this (limited) application....

[Nuuk]

Class D/T amps 'mess' with the phase and I can tell you from first-hand experience that it will make it near impossible to integrate the drivers properly, no matter how much you play around with DSP or any other method of changing the phase output of the signal. Of course if you can't hear the problem then it doesn't really matter!

It's up to you of course but it seems to me that you are already spending a lot of money on the drivers, and if it was me, I would want to use those drivers to their fullest potential.

I'm not sure what the sensitivity of the other drivers is, but had you considered a tweeter with higher sensitivity to match them? As you describe it, the signal is going to be passing through a lot of pots as it is. Presumably the master volume control is in the PC.

The Valab is based on a tried and tested design but is expensive for what you want. I guess you don't do DIY or you could make your own amps to suit what you need. I've no knowledge of the Marantz clone but for the price of either of them, I can find a number of valve amps on Ebay!

[Pallas]

Originally Posted by Nuuk

Class D/T amps 'mess' with the phase and I can tell you from first-hand experience

Do you have any measurements to corroborate that assertion?

Remember the scope here is a midwoofer, so don't show the effects of the output filter at 18kHz and expect anyone to think that's relevant.

[Nuuk]

Originally Posted by Pallas

Do you have any measurements to corroborate that assertion?

Remember the scope here is a midwoofer, so don't show the effects of the output filter at 18kHz and expect anyone to think that's relevant.

I found the problem crossing from the woofer at around 60-70 hz. The phase issue was confirmed to me by a designer of commercial class D amps.

[bcmbob]

NiToNi,

I'm not trying to be a jerk or discourage your approach, but the first question that come to me is - what type/style of music tickles tour fancy? It sounds like you are after a very powerful system, and that might conflict with the level of clarity and detail you seek. Does that sound right?

I am having a ball with the open back Sunflower speakers driven by the new MyRef Fremen Edition LM3886 design. My room is ~ 200 sq ft. and the SPL produced can exceed what is need for full replication of enormous orchestral recordings.

Active crossover projects have drawn a lot of my attention over the past few years (own a dbx 234) but I have hesitated taking the DIY jump due to my questioning whether the work involved results in that much improvement when the music actually reaches my ears. I guess I wont really be able to answer that till I try it myself - right?

Anyway, take a look at the FE thread and a review I did in a chip vs tube session. Here you have the advantage of avoiding the eBay quality challenges and get full forum support from some pretty knowledgeable contributors - some who are currently experimenting with multi-amp active XO systems.

This is admittedly subjective, but you may be aware that a speaker named the "Statements" is highly regarded and combines a ribbon tweeter and an OB mid driver. I asked the designer of the Sunflowers why he didn't use a ribbon and his answer was - he listened to the Statements and many more, but could not hear a significant advantage over a high quality Vifa dome. I actually have all the wood cut for a pair of Statement but have not purchased the internals yet.

So again, can you detail what led you to seek the system you are describing - that's a question not a challenge.

My_Ref Fremen Edition - Beta build/Fine tuning

https://sites.google.com/site/undefi...diy-sunflowers

Statements

[Pallas]

Originally Posted by Nuuk

I found the problem crossing from the woofer at around 60-70 hz. The phase issue was confirmed to me by a designer of commercial class D amps.

In other words, you are willing to provide absolutely no support for your extraordinary claim beyond a sketchy assertion and hearsay, so it can be simply dismissed by any reasonable person as more likely than not a figment of your imagination.

(Though I think there actually may be something to your assertion. It's just not what you think it is, and what is actually makes entirely irrelevant to this case.)

[Nuuk]

Originally Posted by Pallas

In other words, you are willing to provide absolutely no support for your extraordinary claim beyond a sketchy assertion and hearsay, so it can be simply dismissed by any reasonable person as more likely than not a figment of your imagination.

(Though I think there actually may be something to your assertion. It's just not what you think it is, and what is actually makes entirely irrelevant to this case.)

Somebody asked for advice and I offered mine based on my own experience (not hearsay). I didn't realise that this forum had turned into the Spanish Inquisition since I was last here!