Hi All
With my recent frustration with drivers that weren't really suitable for what I wanted I realized that my lack of experience was hampering my ability to make good decisions. So, I've decided to tackle a smaller project first which has completely different parameters. Specifically, I'm building a system to take out in the desert at the end of the summer. This system must be cheap to the point that I can't lose any sleep if none of the drivers survive the trip. I plan to build the cabinets fairly sturdy, there' s not much that the desert can do to cabinets. But the drivers and the amp, that's a different story. To that end, I've chosen a handful of the cheapest MCM drivers. They may not sound great, but, they will make noise. For this particular audience, that's sufficient.
One of the issues with these drivers is that people complain that the measured T/S params are very different from the published specs. So, keeping in mind that I'm very new to this and probably doing something wrong, here's my frst attempt at measuring T/S parameters of the 55-2982 12" woofer.
I'm using LIMP and a diy jig with a 7.65 ohm reference resistor and an NI Audio Kontrol 1 interface and no external amplifier. For these first measurements, the drivers have not been broken in. Also, the diameter is my best guess from measuring and yields an Sd value that compares to another measurement of the same driver in some other thread that is either here or on the parts express forum.
These are the MCM published specs.
Resonance Frequency Fs: 36.23 Hz
DC Resistance Re: 6.3 ohm
Mechanical Q Factor Qms: 2.823
Electrical Q Factor Qes: 0.476
Total Q Factor Qts: 0.407
Equivalent Cas air load Vas: 121.67 liters
Efficiency Bandwidth Product EBP 89.017
Voice Coil Over Hang X-max 7 mm
MCM Audio Select 12'' Die Cast Professional Woofer - 350W RMS | 55-2982 (552982) | MCM Audio Select
These are my measured specs. FS is slightly higher than MCM reports, that this might be expected as the drivers haven't been broken in.
Fs = 42.70 Hz
Re = 6.30 ohms[dc]
Le = 383.02 uH
L2 = 3894.50 uH
R2 = 25.63 ohms
Qt = 0.29
Qes = 0.34
Qms = 2.15
Mms = 77.70 grams
Rms = 9.716533 kg/s
Cms = 0.000179 m/N
Vas = 76.38 liters
Sd= 551.55 cm^2
Bl = 19.739441 Tm
ETA = 1.70 %
Lp(2.83V/1m) = 95.44 dB
Added Mass Method:
Added mass = 34.02 grams
Diameter= 26.50 cm
These seem to model nicely in a narrowed THAM-15. Using the specs as given, they seemed to need more volume and worked better in the full size THAM-15. The mass of the cone is some 20+ percent ligher than a 12PS100, but, a bit heavier than the Dayton 12. I won't be building horns to take to the desert, most likely that will be some sort of bandpass design so that the amp can be very simple and the crossovers less critical. But, I plan to experiment with the second pair and I don't really care if I let the smoke out in the process.
If you buy four, these can be had for about $30. So they're cheap enough to learn some lessons the hard way with. I'll measure the other three tomorrow and probably whip up a better frame for measurement. Right now they are being suspended from the center of my camera tripod.
One of the reasons that I just built a jig to measure with is that I have some odd 12" drivers lying around. I had no idea that it was so easy to get basic measurements. I also realized quite quickly that very slight variations in the physical environment can cause significant variance in the measured params. In other words, my measurements might be total crap, I'll repeat this tomorrow with a better mount.
I'll post more measurements and pictures as I go. Comments, questions, and suggestions are welcome.
best,
gs
With my recent frustration with drivers that weren't really suitable for what I wanted I realized that my lack of experience was hampering my ability to make good decisions. So, I've decided to tackle a smaller project first which has completely different parameters. Specifically, I'm building a system to take out in the desert at the end of the summer. This system must be cheap to the point that I can't lose any sleep if none of the drivers survive the trip. I plan to build the cabinets fairly sturdy, there' s not much that the desert can do to cabinets. But the drivers and the amp, that's a different story. To that end, I've chosen a handful of the cheapest MCM drivers. They may not sound great, but, they will make noise. For this particular audience, that's sufficient.
One of the issues with these drivers is that people complain that the measured T/S params are very different from the published specs. So, keeping in mind that I'm very new to this and probably doing something wrong, here's my frst attempt at measuring T/S parameters of the 55-2982 12" woofer.
I'm using LIMP and a diy jig with a 7.65 ohm reference resistor and an NI Audio Kontrol 1 interface and no external amplifier. For these first measurements, the drivers have not been broken in. Also, the diameter is my best guess from measuring and yields an Sd value that compares to another measurement of the same driver in some other thread that is either here or on the parts express forum.
These are the MCM published specs.
Resonance Frequency Fs: 36.23 Hz
DC Resistance Re: 6.3 ohm
Mechanical Q Factor Qms: 2.823
Electrical Q Factor Qes: 0.476
Total Q Factor Qts: 0.407
Equivalent Cas air load Vas: 121.67 liters
Efficiency Bandwidth Product EBP 89.017
Voice Coil Over Hang X-max 7 mm
MCM Audio Select 12'' Die Cast Professional Woofer - 350W RMS | 55-2982 (552982) | MCM Audio Select
These are my measured specs. FS is slightly higher than MCM reports, that this might be expected as the drivers haven't been broken in.
Fs = 42.70 Hz
Re = 6.30 ohms[dc]
Le = 383.02 uH
L2 = 3894.50 uH
R2 = 25.63 ohms
Qt = 0.29
Qes = 0.34
Qms = 2.15
Mms = 77.70 grams
Rms = 9.716533 kg/s
Cms = 0.000179 m/N
Vas = 76.38 liters
Sd= 551.55 cm^2
Bl = 19.739441 Tm
ETA = 1.70 %
Lp(2.83V/1m) = 95.44 dB
Added Mass Method:
Added mass = 34.02 grams
Diameter= 26.50 cm
These seem to model nicely in a narrowed THAM-15. Using the specs as given, they seemed to need more volume and worked better in the full size THAM-15. The mass of the cone is some 20+ percent ligher than a 12PS100, but, a bit heavier than the Dayton 12. I won't be building horns to take to the desert, most likely that will be some sort of bandpass design so that the amp can be very simple and the crossovers less critical. But, I plan to experiment with the second pair and I don't really care if I let the smoke out in the process.
If you buy four, these can be had for about $30. So they're cheap enough to learn some lessons the hard way with. I'll measure the other three tomorrow and probably whip up a better frame for measurement. Right now they are being suspended from the center of my camera tripod.
One of the reasons that I just built a jig to measure with is that I have some odd 12" drivers lying around. I had no idea that it was so easy to get basic measurements. I also realized quite quickly that very slight variations in the physical environment can cause significant variance in the measured params. In other words, my measurements might be total crap, I'll repeat this tomorrow with a better mount.
I'll post more measurements and pictures as I go. Comments, questions, and suggestions are welcome.
best,
gs
Last edited:
From what I've read elsewhere, the sensitivity is one area where the specs are pure fantasy. I imagine the xmax might be similar.
Now, I hope nobody rushed out to build anything last night based on my measurements above, because, well, they're wrong. That will teach me to wire up jigs in the dark.
I only discovered this after measuring a smaller speaker and found that it had a BL value that was pure fantasy. Here's the updated, less exciting, and almost certainly much more accurate specs.
Fs = 42.95 Hz
Re = 6.30 ohms[dc]
Le = 352.57 uH
L2 = 1417.58 uH
R2 = 19.36 ohms
Qt = 0.64
Qes = 0.76
Qms = 3.85
Mms = 71.11 grams
Rms = 4.984584 kg/s
Cms = 0.000193 m/N
Vas = 82.50 liters
Sd= 551.55 cm^2
Bl = 12.587464 Tm
ETA = 0.82 %
Lp(2.83V/1m) = 92.30 dB
Added Mass Method:
Added mass = 34.20 grams
Diameter= 26.50 cm
Bl is much lower, and the cone mass is slightly lower, closer to the published mass of the cheap Dayton 12. I'll update my models and see how it does now.
On Edit: Now they are much closer to what the published specs predicted, except not as good, and even less suitable for any sort of horn. WinISD gives a reasonably smooth response in a 6 cube closed box and a tolerable response in a 4 cube box with a 3Db down point of 50hz. Alternatively, an 11 cubic foot vented box...ok...I'll just stop there on that one.
Up next...the break in!
Last edited:
Sorry, I tried to just edit the previous post, but the 30 minute limit kicked in. FYI: I think that those sorts of limits on technical forums are silly. But, ok, back to the sims.
The 4th order bandpass tuned to about 55 hz looks tolerable, but this is just a quick sim. It yields about a 5.5 cubic foot box with usable response from 40hz to about 100 hz (give or take). One note? Maybe at least two or three notes.
Now, back to the break in.
The 4th order bandpass tuned to about 55 hz looks tolerable, but this is just a quick sim. It yields about a 5.5 cubic foot box with usable response from 40hz to about 100 hz (give or take). One note? Maybe at least two or three notes.
Now, back to the break in.
After about four hours of burnin on Satuday I get the following. For this, unfortunately, I did not have small enough magnets to get a good balance around the cone. Six magnets, in 3 groups of two, was it so I don't really know how good this is. I'm going to get some lighter magnets and try again.
Fs = 39.39 Hz
Re = 6.30 ohms[dc]
Le = 360.55 uH
L2 = 1323.37 uH
R2 = 19.51 ohms
Qt = 0.62
Qes = 0.75
Qms = 3.57
Mms = 73.76 grams
Rms = 5.107889 kg/s
Cms = 0.000221 m/N
Vas = 94.56 liters
Sd= 551.55 cm^2
Bl = 12.408088 Tm
ETA = 0.74 %
Lp(2.83V/1m) = 91.86 dB
Added Mass Method:
Added mass = 37.42 grams
Diameter= 26.50 cm
So, it seems that break-in dropped the FS slightly, but not much else has moved. Next up is to measure the other three without break-in to see how consistent they are across the lot.
Driver #2
Fs = 43.49 Hz
Re = 6.30 ohms[dc]
Le = 388.97 uH
L2 = 1172.65 uH
R2 = 18.79 ohms
Qt = 0.65
Qes = 0.78
Qms = 3.77
Mms = 73.36 grams
Rms = 5.311367 kg/s
Cms = 0.000183 m/N
Vas = 78.02 liters
Sd= 551.55 cm^2
Bl = 12.721957 Tm
ETA = 0.79 %
Lp(2.83V/1m) = 92.12 dB
Added Mass Method:
Added mass = 37.42 grams
Diameter= 26.50 cm
So, as before, Fs is around 43hz before break-in but all other parameters are fairly close. IIRC, this isn't that far off from what someone else had measured. I'd be pretty much willing to bet that the last two will be close and I'm guessing that this is a reasonable characterization of these cheap drivers. Let's see if I can measure them before my editing window times out, heh!
As far as measuring XMAX, does anyone have a good suggestion? I have a measurement mic on the way and I can try an ad-hoc distortion approach. I've also heard of people taping a piece of folded paper to the center of the dust cap with a black dot and driving the speaker at Fs.
I don't want to let the smoke out doing this, what do you guys suggest as a reliable and safe way to get a decent approximation of the figure?
tnx,
gs
Fs = 39.39 Hz
Re = 6.30 ohms[dc]
Le = 360.55 uH
L2 = 1323.37 uH
R2 = 19.51 ohms
Qt = 0.62
Qes = 0.75
Qms = 3.57
Mms = 73.76 grams
Rms = 5.107889 kg/s
Cms = 0.000221 m/N
Vas = 94.56 liters
Sd= 551.55 cm^2
Bl = 12.408088 Tm
ETA = 0.74 %
Lp(2.83V/1m) = 91.86 dB
Added Mass Method:
Added mass = 37.42 grams
Diameter= 26.50 cm
So, it seems that break-in dropped the FS slightly, but not much else has moved. Next up is to measure the other three without break-in to see how consistent they are across the lot.
Driver #2
Fs = 43.49 Hz
Re = 6.30 ohms[dc]
Le = 388.97 uH
L2 = 1172.65 uH
R2 = 18.79 ohms
Qt = 0.65
Qes = 0.78
Qms = 3.77
Mms = 73.36 grams
Rms = 5.311367 kg/s
Cms = 0.000183 m/N
Vas = 78.02 liters
Sd= 551.55 cm^2
Bl = 12.721957 Tm
ETA = 0.79 %
Lp(2.83V/1m) = 92.12 dB
Added Mass Method:
Added mass = 37.42 grams
Diameter= 26.50 cm
So, as before, Fs is around 43hz before break-in but all other parameters are fairly close. IIRC, this isn't that far off from what someone else had measured. I'd be pretty much willing to bet that the last two will be close and I'm guessing that this is a reasonable characterization of these cheap drivers. Let's see if I can measure them before my editing window times out, heh!
As far as measuring XMAX, does anyone have a good suggestion? I have a measurement mic on the way and I can try an ad-hoc distortion approach. I've also heard of people taping a piece of folded paper to the center of the dust cap with a black dot and driving the speaker at Fs.
I don't want to let the smoke out doing this, what do you guys suggest as a reliable and safe way to get a decent approximation of the figure?
tnx,
gs
Last edited:
Sorry, I was too slow, somebody tell the admins that 30 minutes on a tech forum is just silly.
Driver 3
Fs = 44.85 Hz
Re = 6.30 ohms[dc]
Le = 393.68 uH
L2 = 1268.55 uH
R2 = 19.65 ohms
Qt = 0.69
Qes = 0.83
Qms = 4.06
Mms = 64.02 grams
Rms = 4.439847 kg/s
Cms = 0.000197 m/N
Vas = 84.04 liters
Sd= 551.55 cm^2
Bl = 11.682981 Tm
ETA = 0.88 %
Lp(2.83V/1m) = 92.56 dB
Added Mass Method:
Added mass = 37.42 grams
Diameter= 26.50 cm
Fs is a little higer, Bl a little lower on this one, but still all in the same ballpark.
And finally, the last driver.
Driver 4
Fs = 44.77 Hz
Re = 6.30 ohms[dc]
Le = 387.39 uH
L2 = 1197.45 uH
R2 = 20.14 ohms
Qt = 0.67
Qes = 0.81
Qms = 3.79
Mms = 67.30 grams
Rms = 4.998887 kg/s
Cms = 0.000188 m/N
Vas = 80.23 liters
Sd= 551.55 cm^2
Bl = 12.101042 Tm
ETA = 0.85 %
Lp(2.83V/1m) = 92.44 dB
Added Mass Method:
Added mass = 37.42 grams
Diameter= 26.50 cm
So, I'm going to put one of these in a hastily thrown together sealed box of about 4 ft^3 to see how they'll work as a noisemaker at volume. Any better suggestions?
tnx,
gs
Driver 3
Fs = 44.85 Hz
Re = 6.30 ohms[dc]
Le = 393.68 uH
L2 = 1268.55 uH
R2 = 19.65 ohms
Qt = 0.69
Qes = 0.83
Qms = 4.06
Mms = 64.02 grams
Rms = 4.439847 kg/s
Cms = 0.000197 m/N
Vas = 84.04 liters
Sd= 551.55 cm^2
Bl = 11.682981 Tm
ETA = 0.88 %
Lp(2.83V/1m) = 92.56 dB
Added Mass Method:
Added mass = 37.42 grams
Diameter= 26.50 cm
Fs is a little higer, Bl a little lower on this one, but still all in the same ballpark.
And finally, the last driver.
Driver 4
Fs = 44.77 Hz
Re = 6.30 ohms[dc]
Le = 387.39 uH
L2 = 1197.45 uH
R2 = 20.14 ohms
Qt = 0.67
Qes = 0.81
Qms = 3.79
Mms = 67.30 grams
Rms = 4.998887 kg/s
Cms = 0.000188 m/N
Vas = 80.23 liters
Sd= 551.55 cm^2
Bl = 12.101042 Tm
ETA = 0.85 %
Lp(2.83V/1m) = 92.44 dB
Added Mass Method:
Added mass = 37.42 grams
Diameter= 26.50 cm
So, I'm going to put one of these in a hastily thrown together sealed box of about 4 ft^3 to see how they'll work as a noisemaker at volume. Any better suggestions?
tnx,
gs
Last edited:
Or, for better power handling, you could build a smaller vented box with an Fb around where you want the response to reach down to, with the intent of trimming any peak in the passband and dealing with signals below Fb with DSP on the amp (the Behringer iNuke amps with their built-in DSP seem perfect for this). You'll get both a smaller box and better power handling - win-win IMO.
Hi,
Thanks for the input. I don't quite get the "better power handling." A vented box will go lower and louder with some correction at the cost of increased group delay and necessary eq. Why does it improve power handling? I'm not disagreeing with you, I really don't know. I think that for these first boxes which have a specific purpose in mind, the sealed box has several advantages. 1) it's sealed which will keep dust/dirt/mud out of it, 2) the eq requirements are really minimal allowing me to use old school simple technology to drive them, and 3) for this application, low bass extension is really not desired so I think that the smoother roll-off will be appreciated. I plan to cross these quite high and would probably build them as simple 2 or 3 way cabinets except that presents other undesirable physical problems. Specifically, I need the tops up high but I can't elevate a lot of weight.
At first I was thinking that a bandpass would be the best application, it still might, depending on how distorted these things sound. But, for now, a sealed box with a compromised alignment seems like it will do the job. I'm thinking just over three cubic feet which gives a Qtc of about 0.9. I can make a vented box about two cubic feet but I'm not sure what it gains me, the difference in size between these two is minimal. At 3 cubic feet, the peak in the bandpass is still too high to not require eq.
This is all based on sims, I've never built either box, so maybe I don't understand the impact of my choices on the final product.
Ok, I suspected that. WinISD doesn't give me cone excursion so I couldn't easily verify that. Any suggestions for an alternative simulator?
On Edit: Never mind, winISD Alpha does give that information.
Basically, for the same size, with appropriate filtering, the vented box can take more than twice the power. That said, if I simply look at my target output SPL, both boxes are well within excursion and power limits. These subs are going to be fed with little more than "home stereo" power. 50 watts is probably going to be about the power limit.
But ok, it makes sense to try both as the boxes are almost identical except for a shelf port in one.
Last edited:
So, finally, I made some sawdust today. I built a test box out of the crappiest materials imaginable. The baffle is MDF left over from an organ that I dismantled, so, 12" hole already cut, woot! The rest of the cabinet is cut up, partially laminated, 1/2 particle board that was a set of closet shelves that we hated and have been sitting outside for a long time.
It was mostly an exercise to get to know my tools and to get a sense of how these drivers might sound. The baffle had a cutout for a midrange, which served as a port hole, which I had every intention of trying to play with, but for the time being, I just sealed it off.
I found the track saw surprisingly relaxing to use as compared to the crappy table saw that I had. I also realized that for me, the small makita router was the right choice. One nice thing about the track saw is that it's really quite easy to correct certain kinds of mistakes that I would have normally used the sander or the jigsaw for.
Took me about three hours to clean up the salvaged materials, throw it together, and then another hour or so for playing too loud in my DJ room. In short, it reproduces 60 to 80 hz much better than my small tops, but, it wont' cut it for me for the long term.
Holy jesus is it heavy though, sizewize it's about 19"x19"x22", so like 4.5 ft^3. The dimensions were dictated in part by the existing pieces. The ghetto system is def not going to be made out of mdf or 3/4 ply.
I now have a measurement mic, so I'll try to take some measurements this weekend after I tighten up the box a bit. It looks a bit silly right now with my clamps on it from front to back to stop the unglued back from rattling.
It was mostly an exercise to get to know my tools and to get a sense of how these drivers might sound. The baffle had a cutout for a midrange, which served as a port hole, which I had every intention of trying to play with, but for the time being, I just sealed it off.
I found the track saw surprisingly relaxing to use as compared to the crappy table saw that I had. I also realized that for me, the small makita router was the right choice. One nice thing about the track saw is that it's really quite easy to correct certain kinds of mistakes that I would have normally used the sander or the jigsaw for.
Took me about three hours to clean up the salvaged materials, throw it together, and then another hour or so for playing too loud in my DJ room. In short, it reproduces 60 to 80 hz much better than my small tops, but, it wont' cut it for me for the long term.
Holy jesus is it heavy though, sizewize it's about 19"x19"x22", so like 4.5 ft^3. The dimensions were dictated in part by the existing pieces. The ghetto system is def not going to be made out of mdf or 3/4 ply.
I now have a measurement mic, so I'll try to take some measurements this weekend after I tighten up the box a bit. It looks a bit silly right now with my clamps on it from front to back to stop the unglued back from rattling.
Last edited:
I played with it a bit this morning. I realized that the crossover switch on my mackie was set to 80hz, it sounds a lot better crossed over at 120hz. It seems to take quite a bit of power, but that's just seat of pants based on past experience with my amp and system.
One is stupidly loud in my small house and allows me to really drop the volume on my tops for a much nicer balanced sound. I think that it will take at least a pair to be useful outside. For their intended purpose though, it will work as they're going to be used in a very small area.
I'm pleasantly surprised considering the cost of the driver.
Off the cuff, I'd say that the sensitivity ratings are BS though. How do I measure sensitivity? Also, I still don't know how to safely measure xmax?
tnx
gs
One is stupidly loud in my small house and allows me to really drop the volume on my tops for a much nicer balanced sound. I think that it will take at least a pair to be useful outside. For their intended purpose though, it will work as they're going to be used in a very small area.
I'm pleasantly surprised considering the cost of the driver.
Off the cuff, I'd say that the sensitivity ratings are BS though. How do I measure sensitivity? Also, I still don't know how to safely measure xmax?
tnx
gs
Sensitivity can be using a sine wave of 2.83 volts for an 8 ohm driver. Set the voltage at 60 Hz (where it will be most accurate) then use an SPL meter at two meters laying on the ground outdoors, subtract 6 dB for a one meter equivalent.
You can see the peak to peak excursion quite easily with a white or silver dot on the cone, Xmax is a one way measure, the cone has reached Xmax when the excursion is double the Xmax rating.
Since your cone won't reach Xmax until near full power, make sure you do the tests quickly, and leave time between tests at different frequencies.
You can also just run pink noise, set the HP and LP, run the amp up to clip and see if the speaker exceeds Xmax, if it does, lower the peak limiter settings.
Run of the mill electronics hobbyist test gear: decent, but not spectacular oscope, DMM, function generator. Radio shack sound level meter. Dayton measurement mic and a suitable interface. Most other test gear that I have is of a similar caliber but made for RF measurement.
I build a simple jig for limp.
I suspect that Art's method will be a good start for sensitivity.
I have a dumb question about this. Is this done with the speaker in, or out, of a box? If out, don't you get phase cancellation, if in, doesn't the box affect the measurement?
I'm sure that I'm missing something.
tnx