Hello
I am currently designing my first speaker. It will be a 3-way passive with Seas drivers. As I want the best of all worlds I've decided on a TL (Sub) woofer and sealed midrange for good transient response and decent low end extension. The drivers in question are:
Tweeter: 27TFFC
80 W
91 dB
H0881-06 27TFFC
Midrange: ER15RLY
60 W
87.5 dB
H1455-08 ER15RLY
Woofer: XM-001-04 L26ROY
250 W
87 dB
XM001-04 L26ROY
As you can see, the ER15RLY is quite underpowered compared to the two others. I am thinking it will be crossed at around 150 hz and 2.5k hz. Which means it'll be in the freqency range were it will need to be relatively powerful. Am I right in my two options then: either use resistors on the woofer and tweeter, therefore never reaching the potential of them. Or use a more powerful mid range (larger/two drivers)?
I am currently designing my first speaker. It will be a 3-way passive with Seas drivers. As I want the best of all worlds I've decided on a TL (Sub) woofer and sealed midrange for good transient response and decent low end extension. The drivers in question are:
Tweeter: 27TFFC
80 W
91 dB
H0881-06 27TFFC
Midrange: ER15RLY
60 W
87.5 dB
H1455-08 ER15RLY
Woofer: XM-001-04 L26ROY
250 W
87 dB
XM001-04 L26ROY
As you can see, the ER15RLY is quite underpowered compared to the two others. I am thinking it will be crossed at around 150 hz and 2.5k hz. Which means it'll be in the freqency range were it will need to be relatively powerful. Am I right in my two options then: either use resistors on the woofer and tweeter, therefore never reaching the potential of them. Or use a more powerful mid range (larger/two drivers)?
Using resistors to limit the output is not the answer. That would upset the balance between them.
I think you need to hear from others that have used the ER15RLY that it can go as loud as you need.
I think you need to hear from others that have used the ER15RLY that it can go as loud as you need.
Ok. But if it is at it's power handling limit, the woofer will probably consume 150 ish Watts, and the tweeter something around 15 Watts. If I then turn up the volume anymore, will the ER5RLY burn up/exceed xmax while the two others go louder?
Some is going to depend on the enclosure you use. Either Xmax or power could potentially be a limiting factor. Often one driver is the overall limiting factor. It may not always be the mid, but the question probably should be - is it a problem?
60W is about 17dB. That's 105dB all up. Ok that's a big oversimplification, but that's not too quiet.
60W is about 17dB. That's 105dB all up. Ok that's a big oversimplification, but that's not too quiet.
It is very, very common for rookies to misunderstand a driver's power ratings. All it is telling you is the maximum power the voice coil can comfortably handle. It doesn't mean that that is the power you have to feed the driver. In fact it's much better if you can get the SPL levels that you want and stay well under the driver's max power rating.
To understand how loud a driver will play, look at its sensitivity. For your mid, 87.5dB with 2.83V at 1m (2.83V into 8ohm is about 1W of power). Now every time you double the power, you increase SPL by 3dB. Now just do the math: 1W=87.5dB, 2W=90.5, 4W=93.5, 8W=96.5 ......... 64W=104.5dB at 1m.
Which is pretty much what Allen was saying above. Now in a little more detail, you will lose some SPL's if you listen further away than 1m. In an enclosed room, it's something close to about 3dB/m. So if you are sitting 2m from your speakers, they'll be about 6dB quieter, so about 98dB. If you are sitting about 3m away, about 95dB. Both of which are still pretty darn loud. Loud enough to damage your hearing if exposed for prolonged periods.
But notice that your mid can take more power for shorter periods if you do want them to play louder. Or you also have the choice of doubling the mid up which will gain you another 6dB for the same 60W. Or if that is still not enough, go for a different driver.
Bottom line, your mid choice isn't going to have power problems if you are interested in normal listening levels.
To understand how loud a driver will play, look at its sensitivity. For your mid, 87.5dB with 2.83V at 1m (2.83V into 8ohm is about 1W of power). Now every time you double the power, you increase SPL by 3dB. Now just do the math: 1W=87.5dB, 2W=90.5, 4W=93.5, 8W=96.5 ......... 64W=104.5dB at 1m.
Which is pretty much what Allen was saying above. Now in a little more detail, you will lose some SPL's if you listen further away than 1m. In an enclosed room, it's something close to about 3dB/m. So if you are sitting 2m from your speakers, they'll be about 6dB quieter, so about 98dB. If you are sitting about 3m away, about 95dB. Both of which are still pretty darn loud. Loud enough to damage your hearing if exposed for prolonged periods.
But notice that your mid can take more power for shorter periods if you do want them to play louder. Or you also have the choice of doubling the mid up which will gain you another 6dB for the same 60W. Or if that is still not enough, go for a different driver.
Bottom line, your mid choice isn't going to have power problems if you are interested in normal listening levels.
In keeping with @JReave ... the maximum power rating on most drivers (if honest) relates to safe temperatures in the voice coils on a sustained test signal. Most can handle a fair bit more on properly mixed music, and that can also depend on the genre of music you listen to. So it's not a hard limit, by any means.
That said... you don't want to hook a 60 watt driver up to a 500 watt amp and crank it all the way up. It is absolutely possible to kill it.
At times like this I really yearn for the good old days with power meters on amplifiers so that we can see how hard we are pushing things. Take a look at the scale on this McIntosh power meter...
This is a 600 w/channel amp and notice what's at centre scale ... Yep 6.0 watts. I always get a good laugh when people go, "I sometimes turn it up half way"... In fact most listening occurs at an average of 10 watts or less and even that can be right freaking loud.
So where am I going with this?
Well, if you avoid clipping your amps (as you should) it's likely you won't ever exceed the 60 watt recommendation for that midrange. The only exception would be some kind of techno-rap mashup with 0 dynamic range.
Wthout looking deeply into your driver choices I'd say that combination would be safe for any amplifier in the 10 to 150 watt range.
Last edited:
Well said J.
In a home hifi the power ratings are something you can (mostly) ignore. At least the drivers are all from the same manufacturer the numbers should be consistent — since there is no standard for measuring this specification comparing manufacturer to manufacturer is a can of worms.
I would not, and never do, pay attention to them.
dave
I wouldn't use that sub-woofer in a 3 way unless it was powered. You'll end up with an overall speaker sensitivity of around 82 dB, really low for a 3-way.
If you use this driver, consider a 2 way plate amp with built in DSP and passive crossover between tweeter and midrange.
Otherwise, get a woofer.
If you use this driver, consider a 2 way plate amp with built in DSP and passive crossover between tweeter and midrange.
Otherwise, get a woofer.
Further on erik’s comment. If the is as low as you suggest (<200Hz) it is often better to bi-amp, both cheaper, easier, and better. Passive parts needed at that frequncy get substantiall large and pricey and one has to deal with both the resonant peaks of the mid and the woofer, causing all sorts of grief.
dave
dave
Why is it that the overall sensitivity ends up at 82 dB when all the drivers have higher ratings? Is it the XM-001 that is the problem?
It will be higher than 82dB, and you can already observe that from free space measurement graph.
I think you need to read up on baffle step loss. Briefly, LF's lose 6dB more than HF's when a speaker is in a room away from the surrounding walls. So if your woofer starts at 87dB after baffle step loss it will be down to about 81/82dB. That means you also have to bring the other drivers down to this level as well to get a balanced sounding speaker.
Try Topics 1 and 2 in the Speaker section here - True Audio's Loudspeaker Design Tech Topics Index
And I'll wager this thread is going to be helpful as well - So you want to design your own speaker from scratch!
As the ear is logarithmic in regards to volume you need 10 times more power to be twice as loud.
The speakers really should be comfortable with the power given to them.
You wouldnt drive your car engine on rev limiter all the time.
Actually it's around 85 dB as the SEAS datasheet suggest. SEAS measurements are more close to the reality because it's made on a normal baffle like a loudspeaker baffle and not in infinite baffle like Scan-Speak or other names.
85db is very insensitive.
I use a lot f Fane drivers and they are well into the 90's.
The same amp with better speakers will be much louder for same power.
I use a lot f Fane drivers and they are well into the 90's.
The same amp with better speakers will be much louder for same power.
You are making this unnecessarily confusing. Yes, Seas measures on real baffles unlike just about everybody else but they also show the infinite baffle response too - it's the dotted line at 87dB.
Sensitivity is 87dB as clearly stated in the data sheet. Measurements that include some baffle loss don't change this - the driver sensitivity is still 87dB. The speaker sensitivity however will be lower.
Now why Seas's chart is only showing 2dB with baffle step loss is a bit of a mystery. I'm sure there is a good reason but I just find their measurement philosophy confusing.
Yes, confusing. It looks like it's the response in a complete box. So now the information provided by Seas is insufficient to do a full cross (I looked at other woofers), but easier to produce a useable result by accident.
The infinite baffle curve in Seas datasheets are derived from the T/S parameters, so it's not the actual IB 2 pi response. The measured curve have a slight advantage, because the woofers are box loaded, unlike other manufacturers infinite baffle responses.
Somewhere on the Seas website, there is more info about the boxes and baffles which Seas using for their measurements.
Somewhere on the Seas website, there is more info about the boxes and baffles which Seas using for their measurements.
The way that the infinite baffle measured response is superior, is in its general purpose nature. Once you include a box, it doesn't work for any other and it is difficult to unravel/remove. Also, non-minimum phase information has been included but flattened.
On the other hand it's useful where measurements are not going to be done, or where the crossover procedure needs to be streamlined.
I mean Seas box loaded measurements usually have higher Q than an usual infinite baffle measurement, that is one reason why the Seas measurements are only 1-2 dB lower in level at certain frequencies than the (calculated) infinite baffle response. Okay, the mentioned advantage in sensitivity is not true to the whole spectrum, see the lowest frequency range.
But yes, an infinite baffle measurement are more reliable and usable than the Seas style.
Last edited: