I'm not the expert here, but it seems the first thing to do is determine the TWEETER Crossover, to make sure you can find a bass driver capable of working in that range.
Using an on-line crossover calculator and assuming a 8 ohm tweeter, and given the stated Cap and Coil values on the tweeter, I come up with
2rd Order Butterworth -
- 2800hz -
Capacitor = 5.022 uF
Coil = 0.6431 mH
2nd Order Linkwitz -
- 2000hz -
Cap = 4.975 uF
Coil = 1.273 mH
2nd Ordere Bessel -
- 2200hz -
Cap = 5.182 uF
Coil = 1.002 mH
Given the values stated, a 2nd order butterworth seems most likely.
If we assume Butterworth, that puts the crossover up around 2800hz. We need a woofer, what ever it is, that can function that high.
On the upper bass driver with a single 0.53 uF again assuming 8 ohms, 2400 hz using a 1st order filter, we have 0.52 uF.
That gives us a crossover range roughly 2400hz to 2800hz. Again, we need a bass driver that can function in that range.
Again, assuming an 8 ohm driver, the Low-Bass is crossing at about 625 hz.
That's not a disaster, but it needs to be consider.
Can we assume it is the Low-Bass that are blown or bad?
How much are you willing to spend to fix them?
There is a point where the price gets so high, you are better off scrapping all the drivers, keeping the box and starting over.
In all honesty, while I'm sure this is a nice system, I don't see the drivers as being especially great. So, let's not break the bank. The suggested Visaton at at $22 seem pretty reasonable, but you should really replace all four, though if two is the best you can do, then that is the best you can do. But, if you are replacing all four, then you might as well upgrade the tweeter too.
And if you are replacing everything, then think about a redesign. Mostly of the crossover frequencies.
But then ... like I said, I'm not an expert. But I think first we need to know what it is we are replacing and the job the replacement is being asked to do. Then we need to consider budget. What is an appropriate upgrade considering the rest of the components?
I'm not implying that anyone is steering your wrong. Just saying we need to adopt a broader perspective. In a sense, we are not replacing a driver, rather we are replacing a component in a system, and we need to keep the system in mind, when we select the component.
For what it is worth.
Steve/bluewizard
Using an on-line crossover calculator and assuming a 8 ohm tweeter, and given the stated Cap and Coil values on the tweeter, I come up with
2rd Order Butterworth -
- 2800hz -
Capacitor = 5.022 uF
Coil = 0.6431 mH
2nd Order Linkwitz -
- 2000hz -
Cap = 4.975 uF
Coil = 1.273 mH
2nd Ordere Bessel -
- 2200hz -
Cap = 5.182 uF
Coil = 1.002 mH
Given the values stated, a 2nd order butterworth seems most likely.
If we assume Butterworth, that puts the crossover up around 2800hz. We need a woofer, what ever it is, that can function that high.
On the upper bass driver with a single 0.53 uF again assuming 8 ohms, 2400 hz using a 1st order filter, we have 0.52 uF.
That gives us a crossover range roughly 2400hz to 2800hz. Again, we need a bass driver that can function in that range.
Again, assuming an 8 ohm driver, the Low-Bass is crossing at about 625 hz.
That's not a disaster, but it needs to be consider.
Can we assume it is the Low-Bass that are blown or bad?
How much are you willing to spend to fix them?
There is a point where the price gets so high, you are better off scrapping all the drivers, keeping the box and starting over.
In all honesty, while I'm sure this is a nice system, I don't see the drivers as being especially great. So, let's not break the bank. The suggested Visaton at at $22 seem pretty reasonable, but you should really replace all four, though if two is the best you can do, then that is the best you can do. But, if you are replacing all four, then you might as well upgrade the tweeter too.
And if you are replacing everything, then think about a redesign. Mostly of the crossover frequencies.
But then ... like I said, I'm not an expert. But I think first we need to know what it is we are replacing and the job the replacement is being asked to do. Then we need to consider budget. What is an appropriate upgrade considering the rest of the components?
I'm not implying that anyone is steering your wrong. Just saying we need to adopt a broader perspective. In a sense, we are not replacing a driver, rather we are replacing a component in a system, and we need to keep the system in mind, when we select the component.
For what it is worth.
Steve/bluewizard
Last edited:
I checked the frequency response graphs of several speakers. It looks like the MCM would be best up around 3khz. How it will sound, I don't know, but it responds reasonably well at higher frequencies and rolls off smoother. Most of the others I look at got flakey fast above about 1khz. By the way, it is an $18 speaker being sold for $8.
The VISATON W200 doesn't have a frequency response graph but it is claiming response up to 8000hz, where as the others I saw only claimed response up to about 5khz. That could imply it might do well up near 3khz.
If you could set a working budget range, that would be helpful. Though most of the suggestions do not seem to be too extreme.
Lastly what kind of out come are we looking for? If you are expecting bass down in the 30hz range, be prepared to pay for it. If you just want generally good bass, then things get easier (40hz to 50hz).
Also keep in mind, if you want the low end bass response to go down to deeper frequencies, then that drags the midrange response down with it. A speaker with a 50hz low end might only go to 5khz on the high end, though not functionally so. And a speaker with an 80hz low end might go up to 8khz. But if you want 30hz on the low end, don't expect much more than a rated 3khz on the high end. That high end is also the theoretical max, the speaker may not be realistically function that high.
Steve
The VISATON W200 doesn't have a frequency response graph but it is claiming response up to 8000hz, where as the others I saw only claimed response up to about 5khz. That could imply it might do well up near 3khz.
If you could set a working budget range, that would be helpful. Though most of the suggestions do not seem to be too extreme.
Lastly what kind of out come are we looking for? If you are expecting bass down in the 30hz range, be prepared to pay for it. If you just want generally good bass, then things get easier (40hz to 50hz).
Also keep in mind, if you want the low end bass response to go down to deeper frequencies, then that drags the midrange response down with it. A speaker with a 50hz low end might only go to 5khz on the high end, though not functionally so. And a speaker with an 80hz low end might go up to 8khz. But if you want 30hz on the low end, don't expect much more than a rated 3khz on the high end. That high end is also the theoretical max, the speaker may not be realistically function that high.
Steve
Last edited:
I don't know about unibox, but the stepped woofer crossover is probably an attempt to compensate for the baffle step effect.
How wide is the baffle, it shouldn't be that hard to get an estimate of the Baffle Step Frequency?
Steve
Steve
@Bluewizard.
You raise very good points and I came pretty close to your calc's on the crossover points. of 2500hz for the tweeter and upper woofer. And, 500hz for the lower woofer. That's just me struggling through the numbers though. Please don't ask me to reproduce them! (LOL)
-The baffle is 11.5" wide (although the Black part of the baffle is inset by 3/4").
-My budget is not set in stone. However, I want to stay pretty cheap. Say less than $100 including shipping. Preferably less depending on goal.
-My goal is to really just get them running nominally close to original spec. I know they weren't breaking any technological barriers when new. They will be put into my living room system which is all 70's vintage, so I'll be looking for that 70's coloration. Maybe it's best to stipulate that I'm not looking for upgrades at this time, just to repair or replace the bad components (missing driver) in the most cost effective way without negatively affecting the original sound. Going deep on the bass is not an issue for general listening in the living room.
-Of course, the above statement is relative since I don't really know what the original sound was like. How's that for subjectivity!
-I have 3 of the original woofers. But, like I said originally the speaker tech told me they are rubbing. But for light duty use they might last a while yet. So I could just replace the 2 lower woofers and probably get really close to the sound. However, if the drivers are cheap enough (in the range of those MCM or GRS drivers) I could buy 4 and start by replacing just the lower 2, and trying the uppers as well.
-I might also like to budget in enough to order up some decent quality caps to try out as well (I would be iterative about replacing components rather than shotgun approach). The existing are 8uf film caps of import origin. Say $10 at PE would get me a pair of Jantzen Cross-caps or equivalent.
-Overall my expectations are, I believe, realistic. I have no illusions of these being "audiophile" grade. I have a pair of Acoustat electrostatics with servo amps in my listening room (put together before I decided to go back to school, thankfully) for that.
-Down the road I may decide I like the cabinets enough to start from scratch and build a really good speaker in them. But, I'll cross that bridge when I come to it.
Hope that helps. Everyone has been very helpful with suggestions.
You raise very good points and I came pretty close to your calc's on the crossover points. of 2500hz for the tweeter and upper woofer. And, 500hz for the lower woofer. That's just me struggling through the numbers though. Please don't ask me to reproduce them! (LOL)
-The baffle is 11.5" wide (although the Black part of the baffle is inset by 3/4").
-My budget is not set in stone. However, I want to stay pretty cheap. Say less than $100 including shipping. Preferably less depending on goal.
-My goal is to really just get them running nominally close to original spec. I know they weren't breaking any technological barriers when new. They will be put into my living room system which is all 70's vintage, so I'll be looking for that 70's coloration. Maybe it's best to stipulate that I'm not looking for upgrades at this time, just to repair or replace the bad components (missing driver) in the most cost effective way without negatively affecting the original sound. Going deep on the bass is not an issue for general listening in the living room.
-Of course, the above statement is relative since I don't really know what the original sound was like. How's that for subjectivity!
-I have 3 of the original woofers. But, like I said originally the speaker tech told me they are rubbing. But for light duty use they might last a while yet. So I could just replace the 2 lower woofers and probably get really close to the sound. However, if the drivers are cheap enough (in the range of those MCM or GRS drivers) I could buy 4 and start by replacing just the lower 2, and trying the uppers as well.
-I might also like to budget in enough to order up some decent quality caps to try out as well (I would be iterative about replacing components rather than shotgun approach). The existing are 8uf film caps of import origin. Say $10 at PE would get me a pair of Jantzen Cross-caps or equivalent.
-Overall my expectations are, I believe, realistic. I have no illusions of these being "audiophile" grade. I have a pair of Acoustat electrostatics with servo amps in my listening room (put together before I decided to go back to school, thankfully) for that.
-Down the road I may decide I like the cabinets enough to start from scratch and build a really good speaker in them. But, I'll cross that bridge when I come to it.
Hope that helps. Everyone has been very helpful with suggestions.
Last edited:
I should've also said above that I'm not expecting definitive answers about which drivers will work. I realize an endeavor like this has inherent risks. Just need experienced opinions about where to start.
If you are just replacing the lower bass, then we need only be concerned that they have similar sensitivity to the other speakers, and that they are function just above 500hz to 600hz. That is simple enough. In this case you could intentionally choose low bass drivers. Drivers that sacrifice a bit of midrange for deeper bass.
Then when the time comes, to upgrade the mid-bass, seek out a bass driver that sacrifices a bit of low bass in order to reach up into the mid-range better.
Though it would be wise, to choose both low-bass and mid-bass from the same or similar product lines.
For just the Low-Bass, the field is pretty wide open. But, give some thought to the Mid-Bass when you choose the Low-Bass.
Steve/bluewizard
Then when the time comes, to upgrade the mid-bass, seek out a bass driver that sacrifices a bit of low bass in order to reach up into the mid-range better.
Though it would be wise, to choose both low-bass and mid-bass from the same or similar product lines.
For just the Low-Bass, the field is pretty wide open. But, give some thought to the Mid-Bass when you choose the Low-Bass.
Steve/bluewizard
Ok, so between the GRS (4 for $37) or the MCM (4 for $28) they both look like they could work. And the price is low enough that I'm willing to take the risk. Further down the road if I really felt like putting more into the cabinets I won't be heartbroken over having spent the money on temporary drivers.
Think I'm leaning towards the GRS.
-Despite the hump, I have more confidence in the specs from PE. I noticed the hump lessened when I switched to "heavy fill".
-They have 1.5" VC's.
-The slightly lower efficiency will probably be a better match for the Peerless tweeter, which is either 89 or 91db depending on differing sources for the spec. And, it would be easier to pad the tweeter down than visa-versa. If I only use two new woofers I have no idea what the efficiency of the stock ones are.
BUT, the MCM...
-Model smoother.
-Have better bass response lower down.
-only have 1" vc's. But, that is what the stock woofers are.
-1/2 the shipping vs. PE (not that important).
Somehow my gut is telling me that UNIBOX is not telling me the whole story here. I feel like I'm using a tool that I only have rudimentary knowledge of the basics. Which is not necessarily bad. But, I have the feeling that it can only predict so far what will actually happen.
Think I'm leaning towards the GRS.
-Despite the hump, I have more confidence in the specs from PE. I noticed the hump lessened when I switched to "heavy fill".
-They have 1.5" VC's.
-The slightly lower efficiency will probably be a better match for the Peerless tweeter, which is either 89 or 91db depending on differing sources for the spec. And, it would be easier to pad the tweeter down than visa-versa. If I only use two new woofers I have no idea what the efficiency of the stock ones are.
BUT, the MCM...
-Model smoother.
-Have better bass response lower down.
-only have 1" vc's. But, that is what the stock woofers are.
-1/2 the shipping vs. PE (not that important).
Somehow my gut is telling me that UNIBOX is not telling me the whole story here. I feel like I'm using a tool that I only have rudimentary knowledge of the basics. Which is not necessarily bad. But, I have the feeling that it can only predict so far what will actually happen.
Attachments
Last edited:
Careful you don't end up with too much low and mid and overwhelm your tweeter. Typical consumer woofer sensitivity is ~87-89 db (in parallel 90-92) and your tweeter is maybe 90-92.
Yeah, I'm mindful of that. The GRS is listed as 85db on PE website, which would make me believe the UNIBOX model of 88db for 2 woofers in parallel.
While that makes sense, I wonder if the fact that two drivers in parallel also has an impact on sensitivity.
So, my 8 ohm tweeter is rated at 89 db. If I go with 2-8 ohm woofers that are rated at 85db. But, in parallel impedance drops to 4 ohms are they now closer to 88db?
Or does the parallel impedance not make a difference because the lower woofer is cut off at around 500hz due to BSC?
An amplifier has a more or less constant voltage regardless of load, as long as the load is within the amp's recommended range, for instance 4 to 8 ohms.
When 2 - 8 ohm drivers are in parallel, they experience the same drive voltage as a single one. Because power is V² / R, the two speakers will play twice as loud as one.
Just below the 500hz low pass filter, one driver will be playing at full volume and the other will begin rolling off at 6 db per octave. With increasing frequency, the L3 inductor blocks more and more current to the bass woofer and that 3db gain will shrink to almost 0.
If your drivers have increasing frequency response above 500 hz or if there is a baffle step effect, this may be a good thing.
Speaker building is a trial and error process and if you are a perfectionist you might want to tweak the crossover a little down the road. It starts off slowly and gets more involved. Pretty soon you will be buying Soundeasy and a precision measurement microphone and hundreds of technical books and then you will be setting up an anechoic chamber in the spare bedroom and the wife and kids will leave and then the men in white coats will come for you with a strait jacket. Good Luck ! 🙂
Last edited:
4TORR could you re-read what you wrote in the last post. I think you may have made doe unintentional mistakes. Though I am always willing to be correct if I'm the one making the mistake.
"Because power is V² / R, the two speakers will play twice as loud as one."
They play with TWICE the Power, not twice the loudness. The Loudness gain is either +3dB or +6dB, I can never remember which with out looking it up.
"Just below the 500hz low pass filter, one driver will be playing at full volume and the other will begin rolling off at 6 db per octave."
No ABOVE the crossover one will be rolling off at the crossover slope for that particular type of crossover. It could be 6dB, 12dB, or whatever. BELOW the crossover frequency in a 2.5-way or a 3.5-way system, BOTH drivers will be playing at full volume.
Now I confess there may have been an element to the discussion, that I missed, and I'm willing to be correct, and I also know that these thing tend to be written spontaneously, so error creep in.
So, am I right, or am I missing something?
Steve/bluewizard
"Because power is V² / R, the two speakers will play twice as loud as one."
They play with TWICE the Power, not twice the loudness. The Loudness gain is either +3dB or +6dB, I can never remember which with out looking it up.
"Just below the 500hz low pass filter, one driver will be playing at full volume and the other will begin rolling off at 6 db per octave."
No ABOVE the crossover one will be rolling off at the crossover slope for that particular type of crossover. It could be 6dB, 12dB, or whatever. BELOW the crossover frequency in a 2.5-way or a 3.5-way system, BOTH drivers will be playing at full volume.
An externally hosted image should be here but it was not working when we last tested it.
Now I confess there may have been an element to the discussion, that I missed, and I'm willing to be correct, and I also know that these thing tend to be written spontaneously, so error creep in.
So, am I right, or am I missing something?
Steve/bluewizard
You're wrong, unfortunately. 🙁
Two drivers in parallel give 10*ln(2)/ln(10)= 3db more sound output. Please look it up. We are talking electrically now, and not adding any theoretical acoustic coupling between drivers.
Strictly speaking, the beginning of a crossover's roll-off starts at near 0 hz in the case of an ideal inductor and near infinite hz in the case of an ideal capacitor.
We often speak of the crossover point as the -3db point.
The -3db point is used to describe the "knee" in the curve where the driver is already operating at only half power and is beginning its nominal ascent slope. In fact, if 1 db is the smallest change we can hear, the roll-off starts a full octave before the crossover point !
Are you missing something? Yes, your diagram shows a 3.5 way speaker.
In this thread we are discussing a 2.5 one. 🙂
Two drivers in parallel give 10*ln(2)/ln(10)= 3db more sound output. Please look it up. We are talking electrically now, and not adding any theoretical acoustic coupling between drivers.
Strictly speaking, the beginning of a crossover's roll-off starts at near 0 hz in the case of an ideal inductor and near infinite hz in the case of an ideal capacitor.
We often speak of the crossover point as the -3db point.
The -3db point is used to describe the "knee" in the curve where the driver is already operating at only half power and is beginning its nominal ascent slope. In fact, if 1 db is the smallest change we can hear, the roll-off starts a full octave before the crossover point !
Are you missing something? Yes, your diagram shows a 3.5 way speaker.
In this thread we are discussing a 2.5 one. 🙂
Last edited:
Ok, I've been thinking this over more thoroughly. (which is probably a mistake 🙂
If for the sake of simplicity we assume my crossover points are 2500hz for the tweeter and upper woofer. And, 500hz for the lower woofer (Baffle step correction). Also for the sake of simplicity we leave out the area where the slopes of the crossover points over lap.
Then, above 2500hz only the tweeter is playing and the nominal impedance is 8 ohms and the nominal sensitivity is 89db.
Between 500 - 2500 hz only the upper woofer is playing and again the nominal impedance is 8 ohm. Here I should be trying to match the sensitivity of the replacement woofer to the tweeter. So, should be nominally 89db?
And, below 500hz both woofers are playing in parallel. In this range the impedance drops to 4 ohms. Which means I see a 3db increase due to double cone area, and another 3 db increase due to halve the impedance. Total increase is approx. 6 db for the BSC. However, for the lower woofer, maybe matching the sensitivity is not as critical?
So, since my tweeter is 89 db given a choice between 2 otherwise good woofers, I should choose the woofer with published spec of 89db over the other at 85 db ? Otherwise I will probably have to pad the tweeter down lowering the overall sensitivity of the whole speaker?
If for the sake of simplicity we assume my crossover points are 2500hz for the tweeter and upper woofer. And, 500hz for the lower woofer (Baffle step correction). Also for the sake of simplicity we leave out the area where the slopes of the crossover points over lap.
Then, above 2500hz only the tweeter is playing and the nominal impedance is 8 ohms and the nominal sensitivity is 89db.
Between 500 - 2500 hz only the upper woofer is playing and again the nominal impedance is 8 ohm. Here I should be trying to match the sensitivity of the replacement woofer to the tweeter. So, should be nominally 89db?
And, below 500hz both woofers are playing in parallel. In this range the impedance drops to 4 ohms. Which means I see a 3db increase due to double cone area, and another 3 db increase due to halve the impedance. Total increase is approx. 6 db for the BSC. However, for the lower woofer, maybe matching the sensitivity is not as critical?
So, since my tweeter is 89 db given a choice between 2 otherwise good woofers, I should choose the woofer with published spec of 89db over the other at 85 db ? Otherwise I will probably have to pad the tweeter down lowering the overall sensitivity of the whole speaker?
Make sure you can trust the specs of the drivers you're buying. Also, it's easier to pad the tweeter than buy 4 more woofers.
I did make a mistake...
Two woofers in parallel play 6 db louder. 3db more sensitivity because they draw twice as much power plus 3db more efficiency because their Vas values are additive and the efficiency relationship is Fs³ x Vas / Qes.
Hi,
Yes, you should use nominally 89dB woofers.
The KO10DT was 91dB I recall.
rgds, sreten.
- Status
- Not open for further replies.