Adding a rear-firing tweeter to existing speakers

I have a pair of DIY Amiga MT speakers made from a parts-express kit that I am sure many of you are familar with.
If not they are a 2-way, ported, floor standing using a 7" woofer and the Peerless DX25TG59-04 tweeter.
This is their crossover:


Basically I would like to use them to play around with the concept of a rear-firing tweeter. See if it matches my room and my taste...
But I have a few questions before I go tearing my speakers apart. I am very much a beginner to all this so please correct me if I am making any bad assumptions.

Driver: Duplicate the existing tweeter?

Physically: where to mount the rear tweeter? On axis with the primary front tweeter? Top mount?

Electrically, what is the easiest way to integrate these into the existing crossover? Presumably an L-Pad and a switch to disable them would be a good idea?

Interested in your thoughts as well to this idea but keep in mind its just an experiment.
I would look at a Fostex FT17H horn super tweeter and mount it on a base so that it sits on the top of the loudspeaker enclosure.

That way you can experiment with the front to back positioning of the supertweeter and even have it firing towards a rear wall.

And, of course, you wouldn't have to modify the enclosure in any way!

The super tweeter can be plugged into the loudspeaker enclosure terminals via a 6 dB or 12 dB crossover and an L-Pad as per the examples shown in this link:

P.S. A construction plan of the mounting base is also provided in the link.
Last edited:
  • Like
Reactions: 1 user

If you would rather consider another tweeter, make sure that it has a higher sensitivity than your Peerless (Peerless ~90dB; Fostex 96 dB).

This ensures you will have the headroom to adjust the L-Pad control to give the optimum loudness level (higher or lower than the Peerless), thus giving full scope for experimentation.

Presumably an L-Pad and a switch to disable them would be a good idea?

You'll probably have realised that you don't need a switch along with the L-Pad, as the control can be turned down to level zero if desired.

Since this is your first posting, I'll mention that 8 ohm L-Pad attenuators, like this UK sourced one, are readily available:

Higher power ratings are available, but they are not necessary for a super tweeter since super high frequencies carry very little power.
I did the same with my monitor audio RS6s. The rear port mounting was the same as the front tweeter housing so an easy install. I used a fixed resistor and it took a while to get the value right (a lot more attenuation than I expected). I also added a phase switch as I wasn't sure what would work best. As speakers are close to the wall (600mm) in phase gives the best imaging.
Awesome, even more helpful advice!
Looking at various tweeters available here in the States, many of the small horn-loaded ones are 4 ohm impedence.
Such as the Dayton Audio ST603-4
Can these safely be used with my 8 ohm speakers assuming I also use a 4 ohm L-Pad and an amplifier that can handle less than 8 ohms? Will it affect the design of the auxiliary crossover to have these mis-matched impedences?
Apologies for the stupid questions, realizing this is more complicated than just pluggin in a 2nd tweeter and facing it another direction.
Apologies for the stupid questions

No questions are stupid when you are on an upward learning curve!

The overall impedance of the speaker system won't be altered by adding a 4 ohm tweeter with a 4 ohm L-Pad.

The addition won't upset any decent amplifier.

One point though, the capacitor (and inductor if used) values will be different for 4 ohm than for 8 ohm.

The capacitor value should be doubled and the inductor value should be halved.
The Dayton horn tweeter looks good for the job.

I suggest a 12 dB/octave filter at either 7 kHz or 10 kHz.

Just alter the capacitor and inductor values shown in the Fostex link to accommodate the 4 ohm impedance as I have indicated.

EDIT: Taking a re-think on those component values - see below.
Last edited:
diyAudio Member
Joined 2007
Just a random thought "Ah La" Briggs, if experimenting why not also try an up-firing tweeter? Might need some sort of protective grill.
Although when "Loudspeakers" was published the Wharfedale tweeters were 3inch cones so using any good small full-range r=driver could be used.
Never tried it myself.
You’re going to shift the phase of the existing fwd radiating response in the crossover region…..this will sound pretty bad at varying distances to the front wall. If you reverse the polarity and you can get the speaker out into the room 3ft or more, you have a chance at a semi cardioid response that might sound ok depending on your taste for directivity.

The Amiga has a nearly half octave passband dip of 4db centered at 10khz……it’s a very ‘polite’ sound…….musicians sensitive to upper harmonic undulations appreciate such a speaker while many find this type of tonal balance a bit boring…….sounds like you’re in this camp.
A pre-constructed crossover is probably your best bet as a beginner, as well as cost wise since individual capacitors and inductors can be expensive.
Luckily the individual componets are quite small and inexpensive for similar quality to what was used in the original crossover. Turns out the most expensive parts are the variable L-Pads which I can't even find at 4 ohm. 2 bridged stereo 8 ohm L-pads about the same price as the drivers.
You’re going to shift the phase of the existing fwd radiating response in the crossover region
My understanding is that this shouldn't be a major issue at higher frequencies? If I take Galu's advice and crossover at 10k that is a wavelength of 3mm (and shrinking as I go higher). Even the original 2-way drivers are probably not in plane to that kind of tolerance as they are both mounted flush to the baffle.
To be clear the plan right now is to leave the existing tweeter crossover the way it is so there will be an overlap from 10K up, not just the crossover region. Will be able to adjust the level betweent he two at least.
Anyways its just an experiment. If people can overcome phasing issues with open baffle dipoles seems like there is a chance I can make something interesting with enough trial and error. Or at least learn somthing.
If you reverse the polarity and you can get the speaker out into the room 3ft or more
How do you feel about top-firing? the ceiling is guaranteed to be a set distance over 3ft away. I guess the theory being the more directional horn-loaded tweeter is limited to bouncing off the ceiling and around the room before it arrive so late that being out of phase isn't an issue?
diyAudio Member
Joined 2007
Just spitballing here but why not simply make an "L" shape using a couple of bits of something or even a small stiff cardboard box and experiment?
Take a signal from the speaker posts using some bell wire and play.
I'd be using a cheap but reasonably robust dome tweeter and just a cap in line; 2uF might work
Got any of those HTIAB speakers lying around? Roadkill stuff. I passed a full system on the kerb yesterday but the Boss wouldn't let me stop and pick them up. If I had I'd have another dozen 3" drivers sitting here.