USM-2 DIY Speakers (ND91-4 + ND16FA-6)

I think it will mater at the crossover point the most. We'll see if I have to hook things up in reverse polarity or not.

 

Yeah. For glue I prefer regular wood glue. Only used the Gorilla stuff to bond PVC to MDF. Regular wood glue is awesome and does not expand like Gorilla glue does.

 

Yeah, wiffy made the final decision about that one. I am a bit concerned about the FR implications though. So I might give a go to the woofer crossover later today and test. LOL! I need a small, cheap and nice test amplifier. So maybe not happening today.

I been giving thought to the Yahama R-S202. Amazing bang for the buck at $100. Could use that for testing these w/o having to disturb the family.

 

Does it have enough juice?

I'm all for that one if it does.

EDIT: LOL! Added to Amazon cart.

 

I have a MiniDSP, but it's basically a pre-amp. No worries about it. I may just measure with my Yamaha receiver the way I did the nanosats. I probably will just assemble the crossover and do some stuff inside the box so I can easily remove it and service it.

Not buying the Lepy. Still considering the DTA-1. The deal is, I don't want to have shit in my garage collecting dust. If I buy something is usually for the long run, and while the DTA-1 looks like it fits the bill, I'm not sure it will get any use other than being in the shelf until much later. In the end I may just test with my Yammy and (hopefully) make small adjustments if required.

Thanks guys!

Now figuring out the layout of the crossover. I got mostly 5W resistor parts, small caps, and not so large inductors. But even those are big relative to the enclosure.

Again guys, thanks!!!

 

I'll post the assembled crossover later when I get around it. I have an idea about the layout. I just bought the 24 gauge speaker cable for the drivers. I have 14 gauge speaker wire, but believe it or not, it is huge and bulky for such a small speaker.

 

K. Finished painting the baffle, installing the tweeter (just pressed it on), putting the veneers on the box, staining the veneers, finished the crossover layout and placing one set of components.

However, the things are still drying out and have not put protection coatings to the veneers cuz the stain has not dried out either.

I tested the individual drivers and they produce decent sound. No measurements yet, nor did I tested crossovers. That will have to wait. Crossing my fingers I don't have to do many changes. No changes is better.

Will produce pictures perhaps tomorrow.

Got very pissed off because I fucked one of installed veneers while cutting the edges of another one. There was no power on earth that would let the patching paste cover the slice. In the end I sanded the hell out of that veneer with a 60 grit deal and put another veneer. This shit is hard. Learning has been painful but hopefully things will sound and look good.

 

Will do. Pictures will be w/o polyurethane coatings.

It is interesting that the tweeter mount does not seem to be designed to go all the way in. There is like a little micro face plate there.

 

Pics as promised:





The baffle needs one or two more coats (maybe), and the veneers need polyurethane (a few coats). But it's getting there.

I'll post the cross over 2.5" X 3.5" board later. It looks good. Need to make the holes in the back for connections (again), and then do some subjective and measurement tests with the crossover outside for fine tuning. Hopefully no fine tuning is needed, but who cares.

Veneer wood is Cherry, and so is the stain. Kind of appropriate since these are my first speakers. Build all from the ground up.

 

So did some measurements with crossovers installed temporarily (using a bunch of wire twist-ons).



I'm not sure what's with the small null around 6kHz to 8kHz. According to simulated response there should be a small null between 7kHz to 8 kHz, but I guess it is what it is. The crossover is at 3.5kHz, so I don't think it's that. I will do close mic measurements later.

There is supposed to be a null around 15 kHz, but it is much smaller than in the specs.

I also expected more output from the port. Possible the bunch of wire twist-ons and extra wires where taking out a lot of effective volume from the enclosure. I had a harder time than I though placing the crossover inside with all the temporary shit and that might have cause too much loss in internal volume. If the enclosure looses too much volume, there is not much the port can do. Hopefully once I solder down the crossover there will be more volume for the port to do it's magic. LOL! possible the crossover was blocking the port a bit as well. The crossover is on a 2.5" x 3.5" MDF board (< 0.25" thick). Again, once I make the connections permanent, the cross over should have ample space and things might get much better with the port output. I did some close mic measurements on the port (which I forgot to save) and the port was definitively active. We'll see how things are once all settles down.

Over all I don't think it's too bad though. I don't feel I should make fine tuning. Happy.

I did some quick listening, but not enough to form an opinion. I was looking for total failure types of issues. More listening will be done once the crossover is well put together.

 

I just did a quick check on the crossover values. I assumed PSD-Lite is doing things right. However, the tweeter crossover (ignoring the tweeter impedance and baffle), seems a little high on paper. I will fine tune this actually, but overtime. Maybe a 6 uF cap with a 0.25 mH inductor instead of 4 and 0.15.

Also, the tweeter attenuator resistor seemed a little high on the model (8.2 ohm), but it may actually need to be 10 ohm based on these measurements. We'll see...

The woofer crossover does not seem as off on paper (did some Laplace stuff to check these issues). Which does seem to correlate with measurements.

The port elbow is probably larger than 1" in diameter (two couple the two 1" sections) and may be introducing some issues. I may later try to use a 1" to 3/4" elbow and use 3/4" PVC at the end the port to see if that produces more air out of the port. It may also help increase internal area for the woofer.

There is room for fine tuning if needed. Fun stuff

 

So I went back to the models while soldering crossovers. Here is a small preliminary stream of consciousness:

1. Bass extension: The ND91-4 is capable of quite a bit of bass for a 3" driver. It is advertised as a 3.5" driver, but it really is just 3". However, it needs a bit of room to do this. In a way, it can deliver more bass than similar drivers on equal cabinet volume. I feel that to hit 70 Hz cleanly I needed probably a good 2 to 3 inches more of height, depth, or width all of which I was not going to increase because I wanted a small monitor. The port can't do shit if the internal volume is too small unless one does some active equalization along with an oversized port of smaller diameter. In other words, I feel this is the best I can do with what I wanted to do and learn. I already increased the cabinet size relative the Parts-Express 0.05 ft3 enclosure, and was not willing to do much more. Note that as the volume is increased further these parts loose their appeal, as larger drivers start to fit and perform better.
2. Tweeter: I don't like the ND16FA that much. It's OK I guess, but I expected a more even response. I sort of blame that 6 kHz to 8 kHz null to the tweeter, and while possible to equalize in the analog domain, it means more parts on an enclosure that does not have the volume to accommodate further components.
3. On paper the crossovers by themselves are not 3.5 kHz. However, the response of the components, the baffle, and the driver impedances move things all over which makes this shit an art. I actually derived the Laplace of the crossovers, and plotted their full frequency response. But there are a lot of factors that makes such an analysis very incomplete. I could do my own thing instead of using PSD-Lite. But I found PSD-Lite did a very good job and it's a free publicly available program.

With that, I'll see what I can do to wrap this stuff. I will post the crossover when finished along with final pics and further characterization. Then I will wax lyrical with subjective evaluation while creaming all over my pants.