Unsatisified with the Sound of My Newly Built Amigas - Options?

[audiojerry]

I have really been enjoying my new immersion into building speakers as a way of entertaining myself during this period of Covid isolation. Since December I have built 3 pairs of speakers: the C-Note MT full kit from P.E., Paul's Overnight Sensations later modified to the Overnight Sensations TMM, and now Paul's Amiga. After owning over 3 dozen "high-end" brands including B&W, Dynaudio, ProAc, Avalon, Dunlavy and others, I have been extremely impressed with the sound quality of these diy designs.

After gaining experience in cabinet cutting and assembly, especially the results I achieved with the Overnight TMM, I felt confident to proceed with building the Amiga's. Once built, I played track 8 of the Sheffield XLO burn-in recording for over 100 hours to break the Amigas in. After several hours of listening with various placements of the speakers, I have been unable to arrive at a satisfying experience.

While the bass has been extremely impressive, the overall tonal balance I pick up is a sort of amber coloration and lack of openness. This is in contrast to the cleaner, clearer sound of the Overnight TMM, which also has great bass, but does not reach the lowest depths of the Amiga. With several recordings the amber coloration lended a nice touch to the music, especially with piano and symphonic pieces, but I prefer a more neutral sound.

So now I am wondering what my options are. Is there a way to modify the sound of the Amiga? The cabinets are completely and permanently sealed, but I screw mounted the xover board on the wall directly behind the woofer for easy removal. So I can modify the xover, and replace the drivers. I don't know if the coloration comes from the Dayton RS180 aluminum cone midbass or the Peerless DX25TG59-04 tweet. Personally, I have always preferred the sound of paper cones.

I am more than willing to replace the drivers with suitably sized replacements. I want to use the same 30 l cabinets of course. I found several 6 1/2" and 7 1/2" drivers that could fit, and I am willing and able to modify the cutout if needed. I have not yet researched a possible replacement for the tweeter.

I am wondering if the easiest thing I could try first is using an L-pad attenuator in place of the R2 resistor on the tweeter.

Unfortunately, I am not educated enough in xover mods nor to I have the test equipment to determine what changes I could make to the xover. I am willing to pay or trade for some xover expertise to someone who can offer help. Or, would I be better of trying to sell for parts or sell as-is.

Here are some photos:



I can't put a finish on the cabinets until I can get in the garage when it's warmer.
The speakers look distorted due to the camera lens

[Kornbread]

I'm wondering if a polarity connection might be reversed? You can check the polarity of the woofer with a small 1.5v battery. Touch the input wires to the posts and watch which way the woofers move. Not so easy with the tweeters. I just tap the battery with the input wires, not hold it on there or anything like that.
Of course, if you have a way to measure the impulse/freq response ...

[Blenton]

This is also a good thing to check and is usually where I get a speaker assembled incorrectly. Woofers are easy to tell if one is wired incorrectly because it kills the bass. Tweeters just sound strange when one of them is wire incorrectly. If both are wired backwards, it can cause a huge dip in the response and maybe give that soft muting and coloration the OP describes.

[Blenton]

I'm unfamiliar with the Amigas or their crossover design other than I've seen good reviews of them. However, looking at the schematic of the crossover that you posted with the box sealed up, it looks like the tweeter output between the green 1.5 ohm and 2.4 ohm resistors may be shorting to the white resistor that is part of the woofer network. The white resistor and the green 2.4 are set next to each other, but it looks the top leads may be touching instead of them meeting at the common ground with the yellow twist cap. That would cause some irregularities if they are indeed shorted.

Else, I can only guess at how they should sound. I've used the DX25 in other designs and find it to be very natural and open, maybe even a tad laid back compared to metal domes, but not colored.

[djg]

Your baffle is laid out wrong. You have the tweeter shoved way up in the corner. It should be lower and centered. You modify a design and say it sounds bad. A good start would be to build it as designed, then start modifying if you don't like it. JMHO.

[Unbiasedsound]

This is the first time I agree with you. He should of kept the original design and then modify it instead of going back wards.

[billfitzmaurice]

Offsetting the tweeter wouldn't account for what he's hearing. I wouldn't even venture a guess without seeing at the very least an SPL plot.

[djg]

You know more than I do.

[Unbiasedsound]

Hmmm so you are confident its not the off set of the tweeter but then say you wouldnt even venture a guess without seeing at the very least an SPL plot? LOL

My guess its the off set tweeter. Any time you change a design its going to sound different it dont take a rocket scientist to know this its just common sense. LOL

djg doubts himself without any legit reasoning. We call that nut riding. LOL

[fpitas]

Yeah, a measurement would sure help.

[djg]

On a more helpful note, Paul C. has used that same cab on other designs. The Classix 2.5 uses 2 paper woofers. Two other designs with NLA drivers have been scrubbed from his redone website. The wayback machine could find them I reckon.

[rpb]

I'm unfamiliar with the Amigas or their crossover design other than I've seen good reviews of them. However, looking at the schematic of the crossover that you posted with the box sealed up, it looks like the tweeter output between the green 1.5 ohm and 2.4 ohm resistors may be shorting to the white resistor that is part of the woofer network. The white resistor and the green 2.4 are set next to each other, but it looks the top leads may be touching instead of them meeting at the common ground with the yellow twist cap. That would cause some irregularities if they are indeed shorted.

Else, I can only guess at how they should sound. I've used the DX25 in other designs and find it to be very natural and open, maybe even a tad laid back compared to metal domes, but not colored.

Yep... That sure looks like the problem to me. Really hard to see in the picture, but they appear to be touching, and maybe even soldered. (The 2 ohm, and the 2.4 ohm resistors appear to be touching at both ends.)

[audiojerry]

I believe the xover layout is correct. A different angle shows the 8.2 cap wired to the 2.0 white resistor above and not touching the tweeter circuit. The 5.1 resistor is wired to the + side of the 2.4 resistor.
On the ( - ) side the 2.0, 2.4, and the .35 inductor are all wired together. This photo was taken prior to all final soldering was completed.


This is a link to the Amiga xover layout
https://www.parts-express.com/pedocs...ver-layout.pdf

I have both speakers connected in phase. I'm new at building speakers and xover's, but I know enough about wiring to ground and knowing the difference between + and -.

After my initial posting I have pulled the xover's of both speakers and disconnected the 2.4 ohm L-Pad and am hearing a BIG improvement, but a little too much HF now. I thing a variable attenuator might be
all I need, which will allow me to play around with the tweeter output.

I'm seeing an 8 ohm variable L-Pad from P.E. When I replace the 2.4 ohm L-Pad, should I make any mods when inserting the 8 ohm variable, due to the tweeter being 4 ohm?

[djg]

If everyone liked the same thing, We'd all be building one speaker.

[rpb]

I believe the xover layout is correct. A different angle shows the 8.2 cap wired to the 2.0 white resistor above and not touching the tweeter circuit. The 5.1 resistor is wired to the + side of the 2.4 resistor.
On the ( - ) side the 2.0, 2.4, and the .35 inductor are all wired together. This photo was taken prior to all final soldering was completed.


This is a link to the Amiga xover layout
https://www.parts-express.com/pedocs...ver-layout.pdf

I have both speakers connected in phase. I'm new at building speakers and xover's, but I know enough about wiring to ground and knowing the difference between + and -.

After my initial posting I have pulled the xover's of both speakers and disconnected the 2.4 ohm L-Pad and am hearing a BIG improvement, but a little too much HF now. I thing a variable attenuator might be
all I need, which will allow me to play around with the tweeter output.

I'm seeing an 8 ohm variable L-Pad from P.E. When I replace the 2.4 ohm L-Pad, should I make any mods when inserting the 8 ohm variable, due to the tweeter being 4 ohm?

You would be better off changing resistor values. Also consider the speaker toe in. The geometry is such that the tweeter may be closer or farther away due to the offset.
Try replacing the 2.4 ohm with a 4.7 ohm. That was the xo before the tweeter change. Paul says the new tweeter might be a little bright to some people with the 4.7 ohm. That sounds like what you want.

[djg]

https://sites.google.com/site/undefi...speakers/amiga

[Kornbread]

I have both speakers connected in phase. I'm new at building speakers and xover's, but I know enough about wiring to ground and knowing the difference between + and -.

Well, you're better than me cause I get the polarity switched all the time. .

[Blenton]

I'm in the same boat with you, Kornbread... I even do stupid things like plug the power input to the woofer output, and the woofer output to the power input. Like I just did yesterday... :/

[djg]

Maybe the Amigas don't like your tube amps. Or vice versa. Do you have a little chip amp stashed somewhere?

[billfitzmaurice]

Regarding the tweeter offset, that will affect edge diffraction. Since the distance to the edges isn't the same diffraction will occur at two frequencies with less intensity rather than one with higher intensity. It's a valid design feature, one that I've employed myself. The distance to the top edge will introduce a third diffraction frequency. By all appearances that distance isn't the same as with either edge, which is prudent.

[audiojerry]

Maybe the Amigas don't like your tube amps. Do you have a little chip amp stashed somewhere?

That is possible, although considering the Amiga's low impedence and sensitivity, I would have expected poor bass response not issues with mid / hi frequencies, but the
bass response is awesome and punchy. The Overnights are similarly sensitive, but they sound great with my tube amplification. The amp has 4, 8, and 16 ohm taps.

[Chris Roemer]

1st (and easiest) is to try your tweeters "reverse polarity" (to the way you have them now).
Beyond that, if you want them "bluer"? - (not sure about using colors to describe speaker "coloration" - Oh WAIT ! - I get it? ?) - you COULD swap the position of the tweet's L-pad resistors ... THAT should brighten up the tweeter quite a bit (like maybe +5dB? - use the 2.4n in series, and the 5n across the tweeter).
If they're too bright (blue?), then you'll be happy somewhere in-between there.

Paul says "if you like the Overnight Sensations, you'll like this"; of course, he designed both.