small 3 way. Is this a good idea?

ive wanted to build a 3 way for a while, and i have a surplus of drivers lurking, waiting to be used.

The idea is to use these, if its feasible.

They are: a 6 1/2 inch woofer (good to about 2k and Fs at 40hz), A5 for mids (~600-6k), and ribbon from 6k up.

What do the 3way experts think? Waste of time?

My goals are to control polar patterns, and keep crossover points out of the critical 1-4k area, bass extension with the woof should be 35-40hz vented.
I too have been thinking small three way for the same reasons. My plan is to use the same woofer/tweeter as my best set (Seas ER18 etc.) and trying to keep as many things the same as possible, then do a midrange. First guess is 800, 5K first order crossovers. Is Vanderstein right? Picking the mid has been the issue. I am leaning to the FE89. I have not found any mid I think would be happy higher.

So, is a three way shallow crossover really better in a typical living room over a two-way steep crossover? Can this really nail Joni Mitchel's voice? Can it keep John Williams's guitar from sounding metallic? Harry James trumpet from being edgy?

Let us know what mid you pick.
Good idea as it reduces distortion due to the doppler effect. I would not use a ribbon though there overpriced and the performance is not as good as a ring radiator/dome. I would use the vifa xt-19 and cross ~4k. Don't know what the A5 is.

This is the second time recently the XT-19 has been mentioned over the 25. I am assuming it has the same basic tendencies to be very smooth when used higher? I was also thinking OM-1 put will use the Seas so I can keep as many things the same as the "reference" two ways.

I bet we are thinking about the same thing, but could you describe Doppler effect in the context you mention?
Thanks its nice to know i havent lost the plot!

The woofer is Mission CP168, the midrange is an Alpair5, and the ribbon is a Fountek.

All of these are 'budget', except maybe the Alpair5, though it wasnt expensive. Around £50 per speaker without the cost of the alpair5 (dont recall their price). 1st or low order is the aim, within the xo points mentioned. I own all these already.

Normally Id follow the formula of 8 inch for a 3way, but im using what i have sat here unused.

The CP168 is paper cone smooth natural roll off, slight peaking at 2k and 3k.

The Alpairs I have show a cone edge resonance at bang on 400hz (recently found with test tones during a Zplot).
Im hoping the small alpair is omni enough at the HF xo point to match the ribbon.

Also Im itching to use the ribbon, and this seemed to be the perfect opportunity (or excuse) to use it.

Perhaps a little compromised, but maybe not too bad.
Last edited:
@ Chris661

Yes I agree Chris.

Since the cross over is intended to be ~6k I was musing that perhaps 1st order at 6k would be ok. It may well turn out that 2nd is the lowest order that is practical.

This would likely mean that the Alpair5 would have a matching 2nd order BP, and perhaps only the woofer would use a 1st order LP. In this way a measure of BSC may be applied, since the A5 efficiency will dictate the system efficiency (its the lowest)

I plan to align phase between Mid and HF, and have the inversion on the woofer, provided I can get away with doing that.

However, the NeoCD3.5H I have here are recommended to use a 3rd order at 3k minimum, so I would have around 6db less attenuation (using 1st order), and since Im thinking of using the Alpair, I could move the HF x-o higher still (polar behaviour of the A5 being the restriction)
To be fair the Alpair5 I have are the original model (is there gen2?). They have their flaws, the cone edge resonance mine have is a bummer. Saying that i cant recall if it occurred in a free air or enclosed Zplot, so i should verify when i get time.

Also, id really like to get some decent metal woofers to replace the missions, but the CP168 are cheap and surprisingly clean. Just a tad ugly.

Im sure however, that the newer alpair6 are comparable.

Some may be thinking 'why not just build a FAST?) Well. Nice as these small near fullrange drivers are, they arent as good at HF transiently.

Ive been so impressed by the ribbon (Quoted 95.5 [email protected]). It was padded down to about 85dB, to match the woofers in my mains. Xo'd 3.5k 2nd order, so far from the recommended.
The thing is, padded some 10dB+, they arent 'full of THD'. Thats at modest levels, [email protected]

I wouldnt run it like this forever, i wouldnt want to risk it, but despite that i know they would play clean to higher volume that the woofers.

Its the ribbons reproduction of 'inaudible' freqs that is addictive. I often wonder if a dome with the same VHF bandwidth would also have similar distortion profiles.
Last edited:
I wonder if that is the attraction of the OM1? People seem to love or hate ribbons, so I suspect there is something there to be known.

What I have learned about full range drivers with the Fostex and Fountek is that they rely on breakup modes to give the perception of a balanced HF. So to use them as mids, they need careful crossover and/or notches. Just looking at their specs, I suspect the Mark to do the same. I was successful to a point in doping the Foxtex 125's to kill a couple of the breakup modes, but the total end result was still not the best. I killed one of the breakup modes on the little FE85's by doping the dust cap. But I tried the same kind of thing on a Zaph that worked on the Fostex and it was a disaster. I wonder what the csd on the Mark looks like?
I'm no expert, but I have a question. Why screw around with a 6" woofer on a 3 way?
Even a 2" dome should be able to cross to an 8" or larger woofer.
In my mind, the 6" drivers are a compromise to enable 2KHz crossover points on a 2 way.
All systems are compromises. If you keep the 6" woofer, how much have you gained?

By the way, My brother has an excellent speaker with a foster plainer tweeter, a Dayton RS-52 dome mid ans a pair of Audax aerogel 7" drivers and a 10" passive radiator.

Size. A good 6 can easily reach low enough to be fine for everything but heavy organ and special effects in a reasonable size room under 20L. If one is building a big floor stander, they sure 8 can reach up to 800 easy, 500 or so for a 10 is no problem. The question is of a complex three way to move the crossovers out of the critical vocal range really is an advantage over a two way ALL things being equal. Vanderstein has been selling this for 20 years. And let me assure you, if you think you are going to build something better than a 2Ce, you better think really hard. For me, size and " those tombstones are not going in my living room".

Never liked removable baffles, but that may be a good way to go with the good plywood. I can compare plywood to my MDF Seas set, and then compare two way to three way.
I have been thinking along the same lines to use as my computer speakers.
Using a cheap 6 inch in a sealed box and two of those LG 80mm full-range crossed to the little 12mm Mylar dome in WMTM format.
Still just a thought experiment but I thought doable with relatively simple XO, sealed would give a small box and an F3 of 55Hz with this cheap Jaycar polycone woofer
I have been thinking along the same lines to use as my computer speakers.
Using a cheap 6 inch in a sealed box and two of those LG 80mm full-range crossed to the little 12mm Audax Mylar dome in WMTM format.
Still just a thought experiment but I thought doable with relatively simple XO, sealed would give a small box and an F3 of 55Hz with this cheap Jaycar polycone woofer
DougL: This was the 1st question i was expecting! Basically im trying to build a 3way with the drivers i already have, and in particular, the ribbons. Using the A5 seemed a logical step (since i have these too) and the 6iinch woofers are the largest i have, excepting the 10inch bassguitar amp speakers i have. I also thought that drivers of these sizes would give more capacity to overlap. Also im cheap, and im already hoarding drivers for tomorrow (maybe i should sell some?) To steal Moondogs words, it is a little 'thought experiment'.
TVRgeek- CSD of the Alpairs would be handy i agree. I have to admit im ignorant of Vanderstein, and i should read up. The best ive found is i think at strassacker oq hobbyhifi (may have spelt it wrongly). My german is basic and i had to rely on translate. According to the reviewer the Alpair5 shows peaks at 700hz and 13k on their CSD. It was a handy document, since they had tested ALL the drivers i have, and give info that i couldnt find elsewhere, particularly in the case of the mission woofers.
Going WMTM would be the only way to match efficiencies with these little 80mm LG drivers
The little Audax tweeter is only 58mm across so C2C is fine at around 75mm ( or 3 inches for our non-metricated Yankee members LOL) and it should work with just a 1.5 or 2uF cap in series.
Still a thought experiment but I do have all the drivers and components needed
Do I really NEED a set of computer speakers that can handle 120 watts continuously?? Even as deaf as I am that would be seriously loud at a metre distance.
I could double the box size and go ported with the extra gain and depth,