Details on building the speaker below

[Jeff B.]

Some expressed interest in building the powered monitors I pictured below. Here is the pic again.



Start with the Parts Express RS722 kit. You won't use all of the crossover parts. But that's OK, the kit is still cheaper, and the baffle comes precut, it has pretty much everthing you need in it, and who can't use a few parts to put back and have on hand?

Next pick-up a pair of these 55 Watt biamp plates from MCM (Hey, PE - you guys really need to carry something like this, or a small fullrange amp, or both. Don't make us go to MCM for this stuff )

http://www.mcmelectronics.com/product/50-6277

Cut your holes to mount the amplifier modules

OK. Now after measuring the drivers on the baffle and modeling the 4th order active crossover in the plate amp I arrived at the following response shaping circuit that I placed between the amp and speakers.




The cap on the tweeter was just for protection against any turn-on thumps, but I don't hear any from this amp, so it may not be necessary. I used it anyway though.

The first notch filter on the woofer combines with the driver's response and impedance to compensate for baffle step, flatten the baffle bump at 1khz and give a little lift at 2kHz, which is needed to get a flat summation at the crossover. The second notch (.20mH and 2.2 uF) suppresses the cone break-up. However, I ended up removing it. I found that the 4th order crossover did a good job here all by itself, and I got a very flat summation without using it, and felt the midrange was at a point where I couldn't improve on it anymore anyway. The reason I left it in the schematic was because I know some people don't like to use these drivers without dealing with this area. I didn't find it necessary though.

With everthing done I set the tweeter knob to the max and backed the woofer knob off about 2/3 of the way to get a flat response. These speakers have an exceptional midrange quality, and sound very good across the board. The only lack the deepest bass, but would work well in a home theater system with a sub.

[Paul Carmody]

Nice step-by-step walkthrough! I never realized making a pair of biamped pair of studio monitors could be so straightforward. Thanks once again for your addition to the collective DIY knowledge!

-Paul

[gmilitano]

Jeff,

Very nice work. That amp module looks decent for the price. You would be hard pressed to DIY the amp for that cost.

[Jeff B.]

Nice step-by-step walkthrough! I never realized making a pair of biamped pair of studio monitors could be so straightforward. Thanks once again for your addition to the collective DIY knowledge!

-Paul

All it takes to make almost any two-way work with this amp is the tuning of the parallel notch in series with the woofers, and this is pretty much a must. If you take the response of about any woofer on a baffle and apply a textbook 4th order lowpass to it you will find pretty decent bahavior at the crossover point but a broad peak will exist centered around 800-1kHz due to the baffle diffraction hump lying between the baffle step and the lowpass roll-off. Here's the response with and without the active/passive filters.



The is the key for about any woofer on a small baffle.

Jeff

[dubbreak]

...
OK. Now after measuring the drivers on the baffle and modeling the 4th order active crossover in the plate amp I arrived at the following response shaping circuit that I placed between the amp and speakers.

So for modelling you just used a textbook 4th order LR at 3K and went from there to design the notch(es)?

This looks like a pretty cool project. I'm impressed with how simple the filter/crossover network is.

[Deward Hastings]

You might like the sound even more (and eliminate any question about need for a notch filter) if you spend a couple bucks on the necessary resistors and lower the crossover to 1600-1800 Hz. The RS180 gets quite nasty in HD above 2kHz.. while the RS28 is quite clean in that range.

Putting the baffle step correction in the box is good if you know where the speaker is going to be placed, but for HT use many receivers have a graphic equalizer built in (I'm thinking particularly almost all the Yamahas) that can do that correction for freestanding or against-the-wall without any additional parts at all.

[philiparcario]

Jeff,

Very nice work. That amp module looks decent for the price. You would be hard pressed to DIY the amp for that cost.

It is a great amp for the price. Phil

[Deward Hastings]

It is a great amp for the price. Phil

The 3886 makes a great amp at *any* price, with enough power to drive almost any mid or tweeter. Two of them plus a LR4 crossover for less than $90 ? ? ? The result is cleaner than any amp followed by a passive crossover at any price . . .

[philiparcario]

The 3886 makes a great amp at *any* price, with enough power to drive almost any mid or tweeter. Two of them plus a LR4 crossover for less than $90 ? ? ? The result is cleaner than any amp followed by a passive crossover at any price . . .

But wait there is more a master volume control , a tweeter volume control , a midwoofer volume control ,complete circuit diagram for download and the nicest quality stock two hole plug power cord I have ever seen. That is why I have 5 of them with 1 more for a spare. Phil

[edlafontaine]

PE had one of those amps to look at for awhile. They let it go in the tent sale where I swooped it up. A thread like this may be the ticket for them to reconsider.

[markk]

Does anyone have a schematic for the filter circuit? And practically speaking, how easy would it be to modify the active filter circuit-anyone have a photo of the pc board/inside of these amps?

edit-nevermind. The circuit is posted as a pdf at Jeff's link.

[Jeff B.]

You might like the sound even more (and eliminate any question about need for a notch filter) if you spend a couple bucks on the necessary resistors and lower the crossover to 1600-1800 Hz. The RS180 gets quite nasty in HD above 2kHz.. while the RS28 is quite clean in that range.

Putting the baffle step correction in the box is good if you know where the speaker is going to be placed, but for HT use many receivers have a graphic equalizer built in (I'm thinking particularly almost all the Yamahas) that can do that correction for freestanding or against-the-wall without any additional parts at all.

Technically, I agree with you. I also wouldn't normally cross woofers this size in an MTM over at 3khz either. However.........in this case the midrange is as good as anything I have built. I was really quite surprised at how exceptional it was, as well as the overal tonal balance. I also could not detect any audible lobing issues. I had considered the change to a 2kHz crossover, but after listening to this one the customer and myself both agreed to just leave things as they were. By the way, if I model the response with a 2khz active filter you still have to have a notch filter to deal with the peak at 800-1kHz. I don't see any way around it, unless you use an equalizer like you suggested.

In the passive crossover I designed for this speaker I ended up with a 1.7khz crossover (I did not go with the RS722 crossover, but came up with my own instead). In my small LS3/5a updates I used the RS28a and cross it at 1.8kHz in it. It sounds fabulous there too.

Jeff

[Todd G.]

Jeff, thanks for doing this project. This is a perfect example of why several of us have been advocates for this amp module. If someone would work up some mods to allow a few alternative crossover points, this amp could be a real asset to someone that wants a first class system but dosen't want the hassle of building their own crossover. Anyone could build the simple notch you designed. Thanks again.

[neildavis]

If someone would work up some mods to allow a few alternative crossover points...

We already did that--see this post: http://techtalk.parts-express.com/sh...1&postcount=15
The Excel version is attached. By changing resistors you can get just about any crossover point you like.
50-6277.zip

Something that bothered me about that amp is that according to the schematic the crossover isn't really a 4-pole LR filter, unless I made a mistake, which is quite possible. The tweeter filter is OK: two Butterworth filters each with a Q of .71. However, the woofer circuit has two Butterworth filters each with a Q of 1.12. If this is correct, there will be a noticeable peak on the woofer side. The picture shows the textbook LR filter in orange and the response with Q = 1.12 in red.



A DIY version of this amp is circuit #1 at this page:
http://home.comcast.net/~neilrdavis/Plate_amps/
You can change the crossover of this amp by swapping in different resistor packs. The parts list has a tab that helps you select the right pack. If you want to build this, just upload the PWB file to ExpressPCB. The parts list is probably out of date because the ROHS requirements forced vendors to change a lot of part numbers (this design is about 5 years old), so you might have to do some cross-referencing.

[Todd G.]

We already did that--see this post: http://techtalk.parts-express.com/sh...1&postcount=15
The Excel version is attached. By changing resistors you can get just about any crossover point you like.
50-6277.zip

Something that bothered me about that amp is that according to the schematic the crossover isn't really a 4-pole LR filter, unless I made a mistake, which is quite possible. The tweeter filter is OK: two Butterworth filters each with a Q of .71. However, the woofer circuit has two Butterworth filters each with a Q of 1.12. If this is correct, there will be a noticeable peak on the woofer side. The picture shows the textbook LR filter in orange and the response with Q = 1.12 in red.



A DIY version of this amp is circuit #1 at this page:
http://home.comcast.net/~neilrdavis/Plate_amps/
You can change the crossover of this amp by swapping in different resistor packs. The parts list has a tab that helps you select the right pack. If you want to build this, just upload the PWB file to ExpressPCB. The parts list is probably out of date because the ROHS requirements forced vendors to change a lot of part numbers (this design is about 5 years old), so you might have to do some cross-referencing.

Thank you. I don't know how I missed that before.

[billschulte]

Jeff...just ran across your project and am interested to know if you have or could work out a filter like the above for the Modula MT...would love to try it out with the MCM amps and maybe a Fountek ribbon tweeter...thanks

[Jeff B.]

Jeff...just ran across your project and am interested to know if you have or could work out a filter like the above for the Modula MT...would love to try it out with the MCM amps and maybe a Fountek ribbon tweeter...thanks

My woofer filter just flattens the response and allows the amp's filter to take over for the 3kHz crossover. The cap on the tweeter circuit is there just to protect from turn-on thump, but it is important that you have a cap when using a ribbon so the amp doesn't see a short at DC. That being said, I would use the same filter if I were just substituting the Fountek for the RS28, and then I would adjust the level to fit. By the way, I had to back down the woofers some anyway, so a more sensitive tweeter would work nicely.
Jeff

[philiparcario]

Jeff...just ran across your project and am interested to know if you have or could work out a filter like the above for the Modula MT...would love to try it out with the MCM amps and maybe a Fountek ribbon tweeter...thanks

Go ahead try it at least pick up a pair of amps better yet grab 4 to get the discount. My 5 are still working fine and 3 of them drive a 4 ohm load.

[toddshore]

Jeff, can you add a link to the Dayton RS722 Kits in your first post?

[Paul Carmody]

Jeff...just ran across your project and am interested to know if you have or could work out a filter like the above for the Modula MT...would love to try it out with the MCM amps and maybe a Fountek ribbon tweeter...thanks

?
I don't like being a pooper, but the Modula MT is a passive design, intended to be driven with one amplifier. You can't biamp it. Also, if you swap in a Fountek ribbon, it's no longer the Modula, either.

However, if you wanted to use the driver pair of the RS180 and the RS28a, actively powered by an MCM plate amp, that is probably doable. You'd just have to work within the constraints of the amp, as Jeff discussed earlier in the thread.