PE 299-145 buyout 5.25 PPcone

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Okay, I gave into temptation and bought several of this buyout 5.25 inch polypropylene cone woofer.

Although it doesn't start out all that well, it is going to be great.

Here is the stock performance:

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


This is just the beginning (the drivers arrived just today), but high fidelity quality is a definite possibility. While this modification will most likely end up using a plug instead of a dustcap, a simple cap/plug substitution doesn't help all that much.

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


The real problem is with the cone. This is preliminary, but two glue rings provide substantial improvement:

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


On the things to do list is a three component prefilter/equalizer, an optimized plug shape, and a more precise cone modification series.

All to follow later.

Until then, good designing and good building,

Mark
 
Hate to burst any “wish bubbles” out there, but you cannot take many horrible drivers, stack them into a vertical array and get accurate sound reproduction. If, however, you stack many accurate drivers into a vertical array, the sound will not be much worse than a single accurate driver.

For those that crave the “deal” and want to make multiple driver arrays, I give you modification 2 of the PE 299-145 5.25-inch polypropylene cone buyout driver. With modification 2, I achieve high fidelity quality performance.

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


At $4.94 each, they must be considered a deal. For a couple of dollars more in parts and some simple work, you can have a $4.94 high fidelity quality driver. The more you modify the less the cost per unit. As a bonus, since the Qts is relatively high, you might even get a hint of bass if you use them open baffle.

I do not know how many PE has or how long they will last. If you want to join me in my little project here, buy soon.

Modification two consists of four steps.

Step one is to remove the dust cap. Any sharp pointy thing will do to get under the edge of the dust cap rim and peel it away from the cone. Just pry up with the sharp pointy thing until you have enough of the dust cap rim loose to grab with your fingers or needle nose pliers and peel it off. With only a little care, you can remove the dust cap with no damage to either the dust cap or the cone. It is incredibly easy to get off.

Step two, apply four rings (one non continuous) of Gem Tac glue to the cone and allow to set overnight. The first bead of glue is a one mm diameter glue ring at 5 mm in from the inside edge of the surround. The second bead is two mm in diameter and is located 9 mm in from the inside edge of the surround. The third bead is one mm in diameter and is located 13 mm in from the inside diameter of the surround. The fourth bead is one mm in diameter and is located 19 mm in from the inside diameter of the surround. This fourth bead is not continuous. There are four equally space breaks in the ring; each break is 10 mm in length.

Step three is to insert a three component parallel resonant network in series with the driver. This filter network consists of a .2 mH inductor, a 1.5-microfarad capacitor, and a 24-ohm resistor.

Step four is to install a plug or extension to the pole piece. The plug is a simple cylinder 24 to 25 mm in diameter and 19 mm long. I believe you can buy 7/8-inch diameter dowels at most hardware and hobby stores. Simply cut it to 19 mm in length and using the Gem Tac glue or double-sided tape center and attach it to the pole piece. Most any glue will do, just don’t get the glue into the voice coil gap. You might want to seal the wooden plug with paint or urethane, but it is not necessary.

I plan to continue working with this driver. I envision adding a tweeter and perhaps a simple array of the woofers. All further updates will be posted to my Web site. Give me a couple of days to get the information organized and posted and then look for it either under “driver modifications” or “projects.”

Good designing and building,

Mark
 
Like this I think (it's late here)
 

Attachments

  • screenshot2.jpg
    screenshot2.jpg
    22.4 KB · Views: 1,114
I am responding to the posts and questions generated to date.

If you use the specified glue, expect 10 to 20 years. Just do not get it wet or leave the driver out in the sun. If you leave the driver out in the sun, the polypropylene will not even last 10 years.

How do you know where to apply? In other words, I would like to do my own modifications to other drivers. In sum, it is not easy. I have been working with material vibration modes for over 20 years. Indeed, I co-authored an article on the control of material vibration modes in the Strathern Ribbon driver 20 years ago.

Lacking my experience and understanding, the next best way is to gain access to a functional laser interferometer. Unfortunately, these are very expensive (about .5 million) and costly to rent.

I have not looked at the two illustrations provided, but I have posted photos and illustrations in my MarkMck's Tang Band thread on on my own Web site of other modification projects.

I will include one of my illustrations and a photo on my Web site when I get the page/project up and running.

My thanks to the two people who did post illustrations.

Good designing and good building,

Mark
 
Hate to burst any “wish bubbles” out there, but you cannot take many horrible drivers, stack them into a vertical array and get accurate sound reproduction. If, however, you stack many accurate drivers into a vertical array, the sound will not be much worse than a single accurate driver.

I'm not totally sure i agree with that. I'm no expert, but with multiple drivers doplar distortion will be reduced since each speaker is having to move much less to achieve a given output at a frequency. There was a big discussion on doplar distortion on another thread, so i dont want to start that up again here.

Also, this is just an asumption of mine, but does regular distortion increase with the % of maximum power a driver recieves? ie, the distortion levels (%) for a driver @.25watts will be much less than that driver at 25 watts. Have there been any studies about this?
So, if both of those things are true, ignoring comb filtering, a line array will increase performance of a driver. (this is more of a question that a statement of fact)

also, if I dont want to threadjack so if this post needs to be moved, feel free to.
 
Just a quick comment on Chris8sirhC's point about distortion (and I do not want to begin the distortion discussion again)... Distortion is only one aspect of a driver's performance or accuracy. While I plan to talk about driver distortion as part of project development, it is only one of several characteristics that makes a driver accurate or inaccurate.

While distortion (as it is conventionaly defined) is important, its contribution to the "sound" of a driver is small compared to several other variables.

A horrible driver is going to be horrible for more than just its distortion rating when reproducing a sine wave. Horrible drivers are horrible at any volume level.

As to your second paragraph, an answer would be that it depends. Relations of power, exscurion, spl, acceleration, velocity displacement, and distortion (only one part of accuracy), are too variable to described by a simple rule.

Good designing and good building (and I would recommend the second to any diyer over endless discussions),

Mark
 
Chris8sirhC said:


I'm not totally sure i agree with that. I'm no expert, but with multiple drivers doplar distortion will be reduced since each speaker is having to move much less to achieve a given output at a frequency. There was a big discussion on doplar distortion on another thread, so i dont want to start that up again here.

Also, this is just an asumption of mine, but does regular distortion increase with the % of maximum power a driver recieves? ie, the distortion levels (%) for a driver @.25watts will be much less than that driver at 25 watts. Have there been any studies about this?
So, if both of those things are true, ignoring comb filtering, a line array will increase performance of a driver. (this is more of a question that a statement of fact)

also, if I dont want to threadjack so if this post needs to be moved, feel free to.
Non-linear distortion is decreased with lower excursion, not Doppler distortion. Doppler distortion doesn't exist, according to this article:

http://sound.westhost.com/doppler.htm
 
I guess I just want to concentrate on the in your face problems before getting to the lesser limitations of acoustic transducers while staying away from endless discussions.

Anyway, the first crossover is in the bag so to speak. I have crossed the 5.25 buyout to the Vifa D27. No particular reason for doing this other than I had them and they are a classic tweeter design. Seven components are used.

The combination shown on my Web site is physically offset for transient coherence through the crossover region.

http://madspeaker.com/Projects/299-145Project.htm

In this design test, the drivers are separated diaphragm center to diaphragm center by 8.5 inches. At this spacing, the tweeter baffle is set back from the woofer baffle by 1.25 inches. Both drivers are front mounted (not recessed for flushed) with the baffles.

Good designing and good building,

Mark
 
Mark,

interesting read, those your modification pages.

Am I right in assuming that Gem Tac is essentially the same thing as PE loudspeaker repair glue, i.e. a PVA glue that stays slightly soft?

What are your reasons for using hot-melt glue on the Audax carbon cone and Gem Tac for most of the other modifications? Just the danger of melting the PP cones or also weight or damping considerations?


You might also be the right person to explain the ACD (active cone damping) used by Audio Physik. This has been discussed here and in the Madisound forum several times.

As you may be aware, the Seas L12, L15, L18 cones are aluminum cones with a straight, pretty shallow profile. At the outer circuference, where the cone front is glued to the surround, these cones are bent backwards by 90°, and this bent section points maybe 2 mm to the rear of the driver. I assume this is done do add stiffness to inhibit modes traveling around the circumference of the cone.

Apparently, Audio Physic attach a rubber ring (does not really look like a household rubber band, more like an O-ring seal) to the outside of this ridge. They claim it supresses resonances and lowers third harmonics considerably. I take it this surpresses the first resonance that sits at 4.5 - 8 kHz, depending on the cone size, and hence reduces the amplification of motor nonlinearities by the peak in the FR.

It has been discussed several times, but we have not come to a conclusion whether:
- this is just marketing
- it works through the dension
- by its added mass
- by adding damping to any bending motion


Regards,

Eric
 
Mark, another question to you:

in your mod of the Tang W4, you wrote:
"This one is made from wood. Metal and plastic will work equally well. I have run tests and aluminum and copper provide no measurable improvement in performance because of they act like a shorting ring or cap."

I don't quite understand this, particular the because part. IMHO, there should be some difference in FR and harmonic distortion because a conductive phase plug will lower the inductance of that portion of the VC that extends beyond the pole piece. This will give you more high end extension. It will also change Le variation, whether to the good or bad depends on the exact geometry...
 
Eric,

Thanks for your interest. Just a couple of comments and/or replies.

Not all PVA glues are the same. They have differing adhesive qualities and different densities when cured/dried.

Of the commercial glues I have tested, the GemTac has the strongest adhesive qualities and works best for what I am trying to do to control vibrational modes on polypropylene cones.

The carbon fiber cone does not behave in the same way as the polypropylene and requires a material with differing characteristics to achieve the same results. You can control the vibration modes in polypropylene with low temp hot melt, but it requires a precision of application (particularly bead diameter), that is difficult to achieve and replicate. Because the carbon fiber cone is different, the variable bead diameter is not a problem and the higher density and rigidity of the low temp hot melt is beneficial.

As I have stated before, I am posting glue mods because they are easily duplicated. It requires no special skills to complete the modifications.

I really do not want to discuss other's marketing. As you state, it has been discussed already and I chose not to say anything to those threads.

I have not tested the drivers mentioned and so I do not want to comment on what problems they may have and how effective specific attempts to correct those problems may be. I will, however, say that this edge treatment (bend at extreme outside edge) is not specific to metal cones. As you describe it, the carbon fiber cone has a very similar shape.

Based upon nothing but my understanding of this shape in carbon fiber and my work with other metal cones, the corrections you describe are in the wrong location

I will say, however, that much of what is being done to correct cone vibration modes is only marginally effective (if at all). For example, applying coatings to entire surface of domes and cones (metal, poly, or paper), while it changes the response, does almost nothing for specific vibrational modes.

Lastly, I know people want things to be simple, but just because a theory sounds logical does not mean that it has anything to do with the actual performance of drivers. For example, yes, a non magnetic, conductive metal ring or plug placed in front of the gap will reduce the inductance of the voice coil. The reduction, however, is minor compared to the total inductance of the voice coil. In use, any impact, for distortion or high frequency extension, is swamped by other variables. I can measure the small drop in inductance, I cannot detect changes in performance. And I even doubt the logic of reducing inductance in front of the gap as beneficial.

In sum, I am making changes that can be documented and easily heard as an improvement. I do not wish to stop others from investigating the variable of their choice, but for me it is not.

Again, thanks for your interest,

Mark
 
Just a quick addendum to the voice coil inductance reply. It is important not to overgeneralize theory. There are some woofers with huge voice coil winding overhangs. It is possible that for these woofers, a shorting ring or plug will reduce inductance enough to be detectable in performance. For the drivers I work with, however, it is not a critical variable.

Mark
 
Hi Mark,

thanks for your detailed answers.

Regarding inductance, it makes sense that in those drivers with limited overhand, the change is there, but is too small to have a significant effect. As for larger overhang, I would expect adding a faraday ring just in front of the gap (i.e. not below as well) to be actually detrimental distortionwise: as the coil moves forward, there is less iron in the coil area, hence inductance decreases. With a faraday ring or plug in place, for the same movement, you get more windings that are "shorted out" inductancewise, so you get an even higher Le(x) variation.

As for the Seas aluminum cones, the backwards-bent ridge is always there, and it seems to be working, because the performance of the unmodified drivers is pretty good.

What you are saying confirms my suspicion that it is hard to believe the ACD is the panacea it is advertised to be.



Greetings,

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