Optimus PROX7 Crossover Mod (new thread title)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Regards all,

A new project on the way: upgrading an Optimus Minimus Pro X7 with D26NC55 tweeters, active crossovers, and bi-amplification.

This is an exercise meant to use some of the parts I have collected over the years and otherwise reduce the clutter in my office. Many of you may be able to relate; not being able to pass up a good deal, you wind up with boxes upon boxes of useful (to anyone else, especially the spouse, useless) and interesting audio equipment.

A summary of the items in this project in the order of the estimated date purchased.

8/2001
Alesis Amplifier Module: http://www.apexjr.com/alesis.htm
ALESIS%20AMP%20MODULE%20%232.jpg

Before the term "Chip-Amp" was coined, there was the Alesis "Module". Each board had 10 op-amp filters, LM2875, and LM3886 chips. I bought 10 of these at the great price of $5 each!

03/2004
Optimus Pro X7: http://www.angelfire.com/vt/audio/minimus.html
prox7.jpg

The Pro X7 (Radio-Shack P/N 40-2071) is one of the last evolutions of the famous Minimus 7. The Pro X7 has a roughly 2l ported enclousure with a 4" woofer and 1" textile dome tweeter. The response curve is ragged from the factory with just a cap and coil on the tweeter and the woofer running free. http://murphyblaster.com/content.php?f=minimus.html
I picked up these on Ebay for $20 including shipping!

10/2005
SARTRON 1370-011
1370-011-front-tiny.jpg

This 150W transformer was/is available through several surplus vendors. I got mine for $4.50 each here: http://www.cascadesurplus.com/catalog/product_info.php/products_id/874
Specs: Primaries: 100V, 120V, 220V, 240V, Secondaries: 30V CT 1.0A, 48V CT 2.0A, 30V CT 500mA, and 18V CT 500mA.

06/2007
Vifa D26NC55
An externally hosted image should be here but it was not working when we last tested it.

The manufacturer of this tweeter recently moved production to China, an unfortunate but necessary move to keep up with today's market. I managed to call around and find a pair of the Danish versions, but other than that the value in these lies in that they are just a good tweeter at a decent price ($29 each).

Design
The Pro X7's are the computer speakers in my office at work, but the treble is unsuitable. Dennis Murphy shows in his measurements that this is largely due to the unfiltered woofer and its constructive and destructive interference with the tweeter at and above the crossover point. Rather than redesign the crossover and throw good money in passive components at a cheap woofer and tweeter, I thought it would be a good opportunity to use some of the stuff I've accumulated (and more importantly have a little fun!).

A good tweeter would provide the most significant upgrade in the quality of this design, so that's where we start. The D26NC55 is on top of a shortlist of tweeters that have a faceplate small enough to fit the Minimus enclousures.

The Alesis modules have a "heatspreader" attached to the boards as-is, but require additional sinking for any realistic power output levels. Hmm, the Pro X7's are made of die-cast aluminum. The plan is to incorporate the modules into the diecast aluminum Pro X7 enclousures and use the enclousures themselves as heatsinks.

The transformer(s) would take up too much volume in the enclosures and will have to be mounted in separate power supplies. (Perhaps incorporated into a subwoofer box).

The crossover point on the alesis modules is reported to be at 1.6kHz, 8th order L-R. I do not think that the D26NC55 could handle that low of a frequency, even with an 8th order slope, so the plan is to change out the (surface mount) resistors and capacitors to adjust the filter value up to around 2-2.2kHz, 4th order L-R. This should keep the D26NC55 in its lowest distortion region and still roll off the woofer before it breaks up around 2.5kHz.

Well, lunch break is about over and it's time to get back to real work. Feel free to make suggestions or comments, and check back in the next few days to see the progress!

Regards,
David
 
This project will be worked on during my lunchbreak at work everyday. Here's today's update (07/09/07).

The Pro X7 version of the Minimus 7:
An externally hosted image should be here but it was not working when we last tested it.


Front Baffle Off:
An externally hosted image should be here but it was not working when we last tested it.


Inside the Enclousure:
An externally hosted image should be here but it was not working when we last tested it.


Test Fitted the Alesis Module: (this is going to be tough to fit)
An externally hosted image should be here but it was not working when we last tested it.


Removed Old Tweeter:
An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


D26NC55 Out of Box:
An externally hosted image should be here but it was not working when we last tested it.


D26NC55 Disassembled (Thanks Zaph for showing this on your page):
An externally hosted image should be here but it was not working when we last tested it.


D26NC55 Magnet/Voice Coil:
An externally hosted image should be here but it was not working when we last tested it.


Test fitted the new tweeter: (just barely fits, I would not have known that it would fit were it not for Zaph's page)
An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


Tomorrow I will machine a ring to hold the tweeter in place on the baffle.

Regards,
David
 
That’s it David, you’ve thrown down the gauntlet. Time for a Min 7 bake-off. :)

The first speakers I modded were probably 1978 Min 7s. I still have them, and decided to play one last time.

- cone was heavily damped with lossy silicone many years ago. It lowered sensitivity, but killed the woofer peakiness and added a pleasant bump in the bass (Q = ~1.0 now).

- cabinet walls treated with tar and lined with tar paper. Box stuffed with fiberglass to ~ 1.5 lb/cu ft.

- not wanting to double the budget, replaced the tweeter with the Sonance at Apex Jr. I figured $5 tweeters were about right for this job, and the faceplate is almost a dead on match. Only need to file or dremel the mounting holes a bit to match
http://www.apexjr.com/speakerstuff.html

Here are the system response files, at “on axis” (0deg hor defined here 4 deg below tweeter axis), 30 deg Hor/0 deg vert and 0deg hor/+10 deg vert (this last one is response at standing height in my room). One comment on the low end, the premature roll off is from the windowing, and that 400Hz dip I think is a floor thing, so true quasi anechoic is a bit smoother. The fs is about 120 Hz, which works with the high Q.
http://www3.sympatico.ca/dalfarra/Min7.zip

Xover:
Low pass:
0.93mh iron core (0.24 ohms) with 5.6uF+5ohms “zobel” Zobel is really part of xover, making it 2nd order.

High pass:
2.5uF, 0.93mH(0.24ohms), 8.2 ohms.
An uncorrelated style acoustic xover worked best with these (typical deep reverse notch correlated xovers just didn’t cut it).

Sound: they have a beer budget (Bud) charm with no obvious tonal colourations, no peakiness, etc, and it has that single driver coherence you don’t get in larger systems. Of course its hands are tied by the drivers, and is a bit veiled, but never objectionable or fatiguing. I listen to these for hours and still feel fresh as a daisy. For example the difference between the Audigy card and the much better Julia is still obvious through these, in sound quality and imaging. They do good duty as restricted LF computer speakers driven by an old Bryston and sound much better than your garden variety computer speaker. No real bass but the high Q makes them pleasant enough.

This was a fun challenge, to see what quality I could get out of such a tiny box, for almost no money. They just don’t make them this small anymore, and they visually disappear.

Dave
 
I finished my Minimus 7 upgrade! Thank you DDF for the inspiration. Based on your challenge, the project deviated significantly from the original plans. The alesis modules were saved for later, and a passive crossover approach was taken. Originally I wasn't even going to measure the speakers (tsk tsk). But DDF's results inspired me to set up my measuring rig and rise to the challenge!

Initially, the tweeter was thought to be at fault for the speaker's rising and ragged response. A pair of D26NC55 were ordered and fitted to the Pro X7 housing. The woofer's response was measured to plan the new crossover point, and a large peak at 5k was observed. This led to considering the M11WH09 available on madisound's sale page: http://www.madisound.com/catalog/specials.php

Then I realized, I am about to spend $50 more dollars on these $30 speakers for a total of $130 or so per pair! This is rapidly deteriorating the original purpose of this inexpensive design. So I put the old tweeters back in the PRO X7's and saved the D26NC55's.

The original tweeter was then measured just for comparisons sake. It's not as bad as originally assumed! The woofer was going to need some work though.

The woofer was coated with Mod Podge, 3 coats front, 2 coats rear. The response was measured after each coat and the 5k peak began to disappear. Each time, as the Mod Podge hardened, the peak would gradually come back. The dust cap was then removed and a phase plug was fitted. This didn't help the 5k peak dramatically, confirming that is was most likely a cone breakup node. The dustcap was glued back on inverted, bridging the voice coil to the original point on the cone where the dustcap was attached. This changed the cone modes enough to remove the 5k peak. Below are the response graphs at 1m on axis with the woofer. (The intended listening position.) The ragged one is the original Pro X7, the red one is the measured response of the new crossover and woofer modifications, and the other one is the calculated combined response of the network as designed in Speaker Workshop.
An externally hosted image should be here but it was not working when we last tested it.

The overall cost to me for this modification was $6 for the Mod Podge and $4 total for 2 4.7uF Dayton capacitors. The 1mH inductor is a ferrite core cheapy out of an old Pro LX5 crossover (those speakers are next on the list for mods). The 30uF capacitor was made out of two back to back polarized 47uF capacitors in parallel with the original 4.7uF electrolytic that came in the Pro X7 crossover. The resistors were all on hand.

I can post detailed pictures of the modification process if anyone desires them.

Thanks DDF for the inspiration!

Regards,
David
 
Thanks for the interest. I'm not sure if I still have the pictures, but I can look for them when i get home.

Do you have a pair of Pro X7's that you are trying to mod?

I still use these speakers as my computer speakers and they have very good imaging qualities, though my design process was done somewhat amateurishly in hindsight.
 
Thanks for the interest. I'm not sure if I still have the pictures, but I can look for them when i get home.

Do you have a pair of Pro X7's that you are trying to mod?

I still use these speakers as my computer speakers and they have very good imaging qualities, though my design process was done somewhat amateurishly in hindsight.
Yes, I still have the Pro X-7s, and like you I also plan to use them with my PC. I'm going to change the crossover as you did, and I'd also like to change the woofer cone & voice coil per your modification. I don't have a problem with the crossover, but I don't know how to go about the changes to the woofer. Did you remove the cone with the surround? Also how do you get the voice coil off non-destructively? I probably don't need photos, but I could use a few pointers on working with the speaker cone. Thanks!
Tom
 
I checked my backup drive tonight and I found what I believe to be the original measurement file for the woofer. The file for the tweeter and my crossover file are mysteriously missing, so I would have to re-measure them.

Tell you what, give me a week to find and setup my measuring equipment and I will take a look at these again. I may be able to get a crossover up that doesn't require the modifications to the woofer. At the very least I can get some measurement files that others can take a look at.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.