mixing 4 and 8 ohm drivers

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

If you combine the drivers into a system, it depends on how you combine them.

If they are separated by a crossover, then they are independent.

If they both have the same rated Sensitivity, then they both produce the same output for the same input.

However, what IS the same doesn't always SOUND the same. In general Tweeter, or I should say high frequencies tend to sound louder for a lot of reason. Though of course it depends on how high the frequency is. Baring in mind the highest fundamental note is only about 4khz. With rare and unusual instruments, the fundamental might go has his as 5Khz.

Next, to be able to find a woofer and tweeter that match is rare. Tweeter are typically louder and have to be attenuated to bring them down to roughly the same level as the woofer.

Further different types of tweeters have different output. I have on HORN tweeter that kids out a bit over 90dB. My HORN midrange kicks out about 105dB. My woofer kicks out about 88dB. Cone tweeters are a bit different than cone tweeters which both significantly different than Horn tweeters.

But, my point is, it is rare to get woofers and tweeters of the same Sensitivity, though you can probably find some Dome that are reasonably close to some of the mid-sized woofer.

As to an 8 ohm woofer with a 4 ohm tweeter, you simply design your crossovers using those values. For the Low Pass (woofer) you determine the values using 8 ohms, and for the High Pass (tweeter) you use 4 ohms to calculate the crossover.

But, WHY would you want to do that? If you do, that's fine, it has been done before, but at this early stage of design, I must still ask WHY?

Steve
 
First define "Mix"?

If you combine the drivers into a system, it depends on how you combine them.

If they are separated by a crossover, then they are independent.

If they both have the same rated Sensitivity, then they both produce the same output for the same input.
Since sensitivity is in dBSPL/power, the same sensitivity will produce the same volume with the same POWER input, but amplifiers provide voltage, and drivers pull different amounts of current dependent on their impedance. A lower impedance driver will pull more current, and will be louder when connected in parallel with a higher impedance speaker.

Differences in sensitivity can be "balanced out" with resistors or L pads, but that costs in efficiency, but often good sound quality is more important than efficiency.
 
With the identical 4 ohms speaker and the identical 8 ohm speakers, the 4 ohm speaker might be ever so slightly louder, but neither will be close to the typical tweeter.

Dayton Audio DC130BS-8 5-1/4" Classic Shielded Woofer - 8 ohm
Sensitivity: 88 dB 1W/1m

Dayton Audio DC130BS-4 5-1/4" Classic Shielded Woofer - 4 Ohm
Sensitivity: 88.8 dB 1W/1m

There is your gain.

But, and this is a big but -

Dayton Audio DC160-8 6-1/2" Classic Woofer - 8 ohm
Sensitivity: 88 dB 1W/1m

Dayton Audio DC160-4 6-1/2" Classic Woofer - 4 ohm
Sensitivity: 85.9 dB 1W/1m

Where is the gain from 4 ohms now?

And if what you say is true, then wouldn't it be best if the Woofer were the 4 ohm speaker?

The SENSITIVITY rating is to a Standard to assure that consumers are able to uniformly compare speakers. As long as they can be fairly compared, that's all that matters.

Likely you will have a hard time finding Woofer and Tweeters with equal Sensitivity. So, rather than bother with that, you find the components that best serve your greater purpose and use those, making adjustments as needed.

What I heard you saying is akin to saying, well I want 4 ohm speakers because my amp will put out more power to 4 ohms. NO ... it WON'T. The amp doesn't put out more power, rather the speaker simply consumes more current to no worthwhile end.

Now, if you find some 4 ohm speakers that sound stunning, certainly buy they, certainly use them. But buy them because they sound stunning. The 4 ohm/8 ohm aspect is somewhat irrelevant.

I concede that you might be trying to convey something that I am not grasping. But in general you don't choose a driver simply because the driver is 4 ohms. If it coincidentally happens to be 4 ohms, and it also does a stunning job at what you need done, and it fits your budget, don't not use it. But don't specifically seek out the impedance on the belief that impedance alone provides some benefit.

But, yes, you can mix 4 ohm and 8 ohm, you just design the crossover accordingly. Nothing wrong with that. But there are many more aspects and parameters that far outweigh your choice of impedance merely for the sake of impedance.

Just a few thoughts.

Steve/bluewizard
 
Last edited:
Some clarifications....

First of all, Sensitivity should be rated into 2.83VAC, and then both drivers get the same voltage. For 2.83V/m, this is 1W into 8 ohms and 2W into 4 ohms. Using the average 1W/m designation is not as accurate since both drivers do not receive the same voltage. For 4 ohm drivers the output is roughly 3dB higher than an 8 ohm driver using the 1W/m standard. Since 2.83V and 1W is the same for an 8 ohm driver, but 2W is the level at 4 ohms, you double the power input or go with +3dB to get an equivalent output level at 2.83V.

The term efficiency relates to power consumed vs output, and is generally defined as a percentage. Sensitivity is rated in dB/m at 2.83V as a standard value useful with respect to voltage input.

So- BW- your comparison at 1W/m is invalid, as it is inaccurate.

Your statemnt of design a xover for a set nominal impedance value is also incorrect, as resistance varying with frequency equals impedance. You cannot use a simple equation and come out with the proper values. Adding in Freq Resp, you pretty much need to start from square one. Using this method of creating a xover is not the best bet. And as to using a driver with different impedance from the other- it's not an issue; ever. You just design the xover to compensate. Most of the time, xovers have to attenuate the tweeter anyway, and you can add resistance back at times or leave it as is.
There are many factors involved in a good design.

Bill-
If you have an 8 ohm woofer and a 4 ohm tweeter or full-range that have the same output at 2.83VAC/m, then to either side of the xover (barring room interactions, or even diffraction and baffle-step), they should have the same output nominal level.

BTW, Bill- love that Lancia Stratos in your avatar there...ca 1973?
Wolf
 
The same power vs the same voltage is something of a debate as to which is more fair. At various times I swing to each side.

But, since both these drivers with be in the same cabinet presumably making up the same speaker system, then they have no choice but to always receive the SAME source signal voltage. So, in this case, same source voltage becomes the more fair test of how the speaker will work together.

Again, I'm not saying don't use 4-ohm, I'm simply saying do use it simply because it is 4 ohm.

As to my statement about crossover design, I think we need to give some thought to the likely skill level of a person who would ask such a question. It is not likely that he is going to be building Engineering Degree crossovers, far more likely he will start with basic text book design, and expand and improve as his knowledge base and skill level improve. No offense to the original poster, we were all where you are when we started out.

I also suspect those among us who consider themselves very good crossover designers would be the first to tell you they don't know half of what they would like to know.

Steve
 
Last edited:
The same power vs the same voltage is something of a debate as to which is more fair. Steve

I'm sorry, but this is false. Wattage is a big misnomer in speaker building and is almost a worthless parameter in terms of designing. In terms of relating total output of a completed design with its sensitivity and wattage input is about the only place of usage, aside from knowing when the driver will go into thermal meltdown.

If you use the same power on a woofer and tweeter with same dB/W rating, conjugates to make their impedances resistive, and both are the same nominal impedance, and you have an infinitely wide and tall baffle placed in an anechoic room....

This removes cabinet issues, room interaction, impedance differences, and input differences to the point of being nill. IF AND ONLY IF the drivers have a flat frequency response, will they then put out the same level of output consistently to be rated as the same.

If any of these things change, then your thought that Power/dB is equal will effectively dissolve.

If an amplifier puts out power, it has to put out current or voltage. Solidstate amps are voltage-gain based, and toobs use current-gain. If you base this on voltage as an average av receiver supplies, then you require equal voltages to get a reasonable response. There is no way around this.

Later,
Wolf
 
Bill-
If you have an 8 ohm woofer and a 4 ohm tweeter or full-range that have the same output at 2.83VAC/m, then to either side of the xover (barring room interactions, or even diffraction and baffle-step), they should have the same output nominal level.

BTW, Bill- love that Lancia Stratos in your avatar there...ca 1973?
Wolf

Thanks guys! I'm wanting to match the alpair 6p in a shallow waveguide to a suitable paper woofer(1st order, 550hz x/o). The 86dB Alpair has quite a healthy output in the upper treble(too much?), so the flare would be quite shallow so as to match a woofer with about 90db sensitivity and also not forgetting bsc etc

to keep the cab quite slim, the 3 woofers shortlisted are the 8" Peerless, one of the SS 7" 4 ohm Revelator's and the Vifa NE225 4 ohm. The Peerless is much cheaper and altho it states 8 is rated at 5.8 ohm or something. The smaller Scanspeak surprisingly beats the others on extension

Wolf- Bertone Stratos- not sure the yr, thinking 1970/71? one of my fav designs :)
 
Last edited:
And, what are you talking about? All drivers, regardless of their nominal impedance values, have (or should have) sensitivities based on a giving them a 2.83V RMS input and measuring their generated SPL in dB at 1 meter distance. Your examples clearly show different input voltages required to create sensitivity ratings based on 1-watt inputs for different nominal impedances, but that's all "old school" left over from the a different era in amplifiers and drivers. Some drivers are rated both ways, with a 2.83-volt input and a 1-watt input, however.
Paul

what are you talking about? only 8ohm drivers are measured at 2.8V
ohms law. square root of P * R
ie: 2.83V/8 ohms, 4V/16 ohms, 2V/4 ohms
 
Drivers are specified for sensitivity either @ 1W or @ 2.83Vac
The manufacturer usually tells the customer which they are stating.

Dishonest retailers will use the number that gives the highest apparent sensitivity using any loading case without stating what they are quoting for. These are the ones to avoid.

An example that works in the reverse direction is for some Tannoys that are effectively 10ohms. Tannoy will state 95dB/W @ 1m for the passband. They state up front what the spec is. But we have to realise that to get 1W into that speaker, the noise voltage in the passband has to be slightly higher than the usual 2.83Vac that would be assumed for an 8ohms driver.

The real cheats are those that sell 4ohms and lower drivers who will use the 2.83Vac sensitivity but not tell you, i.e. simply state: sensitivity 86dB
 
Last edited:
And, what are you talking about? All drivers, regardless of their nominal impedance values, have (or should have) sensitivities based on a giving them a 2.83V
maybe you should read ohms law again? nothing else than 8ohms should be applied 2.83V period.
when it says 1 watt it IS 1 watt ! no respected speaker manufacturer on the planet would think of doing otherwise.
 
I said -

"The same power vs the same voltage is something of a debate as to which is more fair. "

To which you said -

I'm sorry, but this is false.

Why do I say there is a debate? Because I've read and been involved in several of these debates. The fact that some people rate speakers at 1w/m and some rate speakers at 2.83V/m again pretty much confirms the debate.

Most often I come down on the side of Voltage, because the signal voltage coming from a given amp driving a given set of speakers, is fixed for all components within the speaker system, ignoring any attenuation for specific drivers. The fact remains, in most cases, one amp is feeding all drivers, and the voltage from that amp is the source that is being applied to the drivers.

Just so the OP is clear about what we are talking about, let me illustrate.

The formula to find the Voltage knowing the Resistance and the Power is -

E = SqRt(P x R) where SqRt() is Square Root.

So, what voltage is applied to an 8 ohms speaker to produce ONE Watt?

E = SqRt(1 x 8) = SqRt(8) = 2.83 Volts

Now the same for a 4 ohm speakers to consume ONE Watt?

E = SqRt(1 x 4) = SqRt(4) = 2 Volts

So, how much power do we get if we apply that same 2 Volts to an 8 ohm speakers, just to give us some perspective on the difference.

P = E² / R = 2² / 8 = 4/8 = 0.5 Watts

Which is why I said, that most often I come down on the side of voltage; all speakers should be driven with the same fixed voltage AC Signal so we can know how they will compare when place together is a box and driven at the same time by the same amp.

The goal isn't to know how a given driver will perform on its own, but how it will perform relative to other drivers. But like it or not, Drive Voltage vs Drive Power is being debated.

Handy Ohm's Law Pie Chart -

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



Wattage is a big misnomer in speaker building and is almost a worthless parameter in terms of designing. In terms of relating total output of a completed design with its sensitivity and wattage input is about the only place of usage, aside from knowing when the driver will go into thermal meltdown.

I'm not sure misnomer is the right word, but I think we still understand what you are saying in context.

Wattage is not worthless, but the key is to select components that result in a complete speaker system that has a predictable Power Handling capacity that is within the range you need.

But Power Distribution is not uniform across the spectrum. For Full Orchestral music, the peak power band is in the range of 250hz to 500hz. Above 500hz the power demands drop quickly. Below 250, they drop, just not as quickly but it still drops pretty quickly.

If the range 250hz to 500hz is consuming 10watts, then the range of 125hz to 250hz is only consuming 4 watts. Conversely, the range of 500hz to 1000hz is consuming a modest 2 watts.

If you are crossing Bass-to-Mid at 300hz, then your Mid drivers are going to need most power handling and excursion range. If you cross Bass-to-Mid above 500hz, then the Mids need not handle quite as much power nor will they need quite as much excursion capability.

So, as an example, if you cross Bass-to-Mid at 800hz and you start with a 100w woofer, then theoretically you only need a 20w mid-range and at 10 watt tweeter. Though I think in practice that would be cutting it to the thinnest of razors edge in terms of power handling. But it would not be unrealistic to have a 100w woofer, a 50w mid-range and a 30w tweeter and feel comfortable calling this a 100w speaker system.

Handy Power Distribution Chart -

354472d1371324139-4-way-speaker-help-high-end-powerdist-.jpg



...

This removes cabinet issues, room interaction, impedance differences, and input differences to the point of being nill. IF AND ONLY IF the drivers have a flat frequency response, will they then put out the same level of output consistently to be rated as the same.

If the various drivers are initially rated the same using the same method and standard, and if they are flat in their working range within a given design, then the output level will likely be the same.

...

If an amplifier puts out power, it has to put out current or voltage. Solid-State amps are voltage-gain based, and toobs use current-gain. If you base this on voltage as an average av receiver supplies, then you require equal voltages to get a reasonable response. There is no way around this.

Later,
Wolf

I think in reality all Audio Amp are Voltage devices. The Amp provides a signal Voltage to the speaker, and the speaker consume Current; just as your car consumes fuel to produce power. When we apply voltage and current is consumed, we can rate the amount of work done as a result in the form of Watts.

On this we agree, but there is still debate as to which is more fair - Sensitivity rated by Voltage vs Sensitivity rated by Power.

It seems to me that we can't know how a given driver will perform relative to other drivers inside or outside the same box, unless they are all fed the same signal. But not all people agree, as can be seen by the fact that some people rate all their speakers driven by 2.83 volts and other drive all their speakers with 1 watt of power.

But my underlying point is, you don't choose a 4 ohms driver simply because it is 4 ohms, in some other way it must serve the design. It must have more significant advantages in other aspects of its performance to be a worthy driver, and if it has all these other advantages and happens to coincidentally be 4 ohms, then it is probably a worthy consideration.

But the fact that a driver is 4 ohms must serve the design in some other way besides the mere fact that it is 4 ohms.

And that is all I'm trying to say.

Steve
 
Let's go back to the original questions -

"say u mix an 8 ohm bass and a 4 ohm full range or tweet, will output be the same if the sensitivity rating is the same for each? or will the 8 ohm driver have less output when used with the 4 ohm ?"

As you can see from the discussion, the answer to your question hinges NOT on the numbers being the same, but on the Rating Standard being the same.

More specifically, both drivers have to be rated at 2.83V for the Sensitivity number to have any comparable value.

Can we come up with a consistent and predictable way to adjust the Sensitivity rating of two drivers rated at 1w/m, so that we gets some sense of the Sensitivity at a fixed input of 2.83v?

If the power is doubled we can simply add +3dB, but how do we deal with the areas in between.


If a 4 ohm driver is rated at 1w/m and an 8 ohm driver is also rated at 1w/m, and they are both rated at 90dB, how can we adjust those number to determine the output with both drivers driven with 2.83v?

In general, a 4 ohms speaker driven with 2.83V will consume 2 watts, so is it as simple as adding +3dB to the 1w/m numbers?

I think more than anything a method for correcting the Sensitivity number to make an equal comparison would be the most valuable tool.

Steve
 
Last edited:
The handy power distribution chart is good at helping to size the heatsink of a narrow band amplifier.
The chart is no good at predicting the peak SPL in any frequency band.

The Chart is a broad average, no it can't predict the power distribution of any one given musical work. I suspect the power distribution between "Bass I Love You" and John Denver's "Rocky Mountain High" are very different.

But the tests were done with full Orchestral music, which has pretty substantial dynamics and wide frequency range, and it servers as a fair average of power distribution across a wide spectrum of music.

The chart can help you have an understanding of the general power distribution of real music, but no it can't predict the power distribution of a single give musical composition. Nor can it predict the peak SPL of a given work of music. But then, it is not intended to do that. It merely gives the power distribution on average of Full Orchestra Music.

Steve
 
I can read and understand Ohm's law quite well, thank you. The point, which apparently completely escapes you, is that the only way to compare the sensitivity of two drivers with different impedances is via the same voltage input because that automatically accounts for those impedance differences and accurately tells you how much SPL the drivers will put out under absolutely identical conditions, meaning input, which a 1-watt input does not and cannot.
Paul

maybe you should read ohms law again? nothing else than 8ohms should be applied 2.83V period.
when it says 1 watt it IS 1 watt ! no respected speaker manufacturer on the planet would think of doing otherwise.
 
Especially in Internet discussion, when ever anyone speaks in Absolutes, they are invariable wrong or wrong within a context.

If two drivers are in the same box and are part of the same speaker systems, then the sound source voltage is being applied to that Speaker is what it is. It is not one voltage for the 4 ohms driver and another voltage for the 8 ohm driver.

To be able to fairly compare these Drivers to know how and whether they can be used together, we need the same average voltage applied to them.

However, if we consider the Drivers independently, an argument could be made for them to all have the same power applied - 1 Watt.

Since there is no consistent standard, we do the best we can with what we have.

I would suspect, if both an 8 ohm driver and an identical 4 ohms driver are driven with 1 watt, we would need to add +3dB to the 4 ohm to get a best guess as to how it would preform in the same application as the 8 ohm driver.

However, my point is, it is best to hedge your statements. To state absolutes is always a sign of fanaticism, and that is never good. Yes, I agree there should be one consistent uniform world standard, but there isn't, and we do the best we can with what we have.

Don't mean to offend, just making an observation.

Steve/bluewizard
 
Last edited:
They are in parallel, and you do the calculation for two resistors in parallel. It's the reciprocal of the sum of the reciprocals of the impedances, or 1/(1/8+1/4) = 3.08 ohms. This impedance is too low for many amplifiers, and may cause excessive heating or burnout of the amplifier at moderate to high volume level. With biamped or a separate amplifier for each speaker, each amp would of course "see" the load of the speaker it's connected to.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.