Resistors and Impedance Matching

Status
Not open for further replies.
After reading every crossover primer, and builder guide I can find, I have yet to find a detailed explanation on how to use resistors to match impedance levels of drivers in a multi-way system. They all just mention that is what they are used for, and one very lengthy site mentioned they do not affect frequency response (although I did notice it was mentioned in the AR kit a resistor was used with an inductor to eliminate a frequency dip so... :scratch: )

Anyway -

I have a 4ohm driver and a 6ohm driver, that I want to use together to form - a 4ohm system nominal impedance.

I already have my crossovers and zobel designed, but I need help with this one last bit to complete the circuit.

Is it as simple as using a 4ohm resistor, in series, on the 4ohm driver to make it an 8ohm circuit, and using a 2ohm resistor, in series, on the 6ohm driver to make it an 8ohm circuit - then paralleling the circuits to get 4ohms?

Good grief, if it is this easy, no wonder they dont mention it.

If so, then question: Do I place the resistor first, before any of the other x-over components, or last, after the zobel, etc.?

Or if Im totally wrong about it all, please disregard and advise.

Thanks.
 
It's really only practical to use resistors to change the impedance of the tweeter (in a 2 way system at least) - and even then, it also serves to pad the sensitivity of the tweeter down to that of the woofer.

The high pass and low pass sections of a crossover do not need to see the same impedance - it is just more simple for them to have the same impedance, and provides a more constant load to the amplifier.

It's not quite as simple as what you are trying to describe here:

"Is it as simple as using a 4ohm resistor, in series, on the 4ohm driver to make it an 8ohm circuit, and using a 2ohm resistor, in series, on the 6ohm driver to make it an 8ohm circuit - then paralleling the circuits to get 4ohms?"

This would actually give an 8 Ohm load - at low frequencies, the high pass section is effectively an open circuit, and at high frequencies, the low pass section is effectively an open circuit, they don't parallel like resistors, that is the whole point of the crossover.

It would be a wasteful way to design a crossover anyway - you're wasting almost half your power in resistors (which will also need to be pretty beefy to dissipate that sort of heat). Depending on whether your woofer is the 4 ohm or the 6 ohm driver, I would try and make the speaker the same impedance and change the woofer - but you will still need to take the sensitivities into account.

These things should be taken into account from the start - don't just go about designing everything else and leave it as an afterthought.
 
Stu said:

These things should be taken into account from the start - don't just go about designing everything else and leave it as an afterthought.

Well, I guess thanks for telling me I was doing everything wrong, but I was hoping to get advice on how to do things right.

FYI - This isnt just a shot in the dark with drivers that fell from the sky; background: I have a McIntosh MSS650 Mobile Component system that uses their 4ohm, 89db, 6.5" woofer, and a 6ohm tweeter made by Vifa. I really like the tweeter, and Vifa is now selling it under their own name. So, I thought I'd give a shot making a component system for my car using that tweeter, and a good woofer, with similar sensitivity. So I picked the Morel MW168, 4ohm, 88db. The McIntosh system integrates the drivers in their system very nicely with a 4ohm system impedance, so - why can't I? Here are the drivers I'm using.

Woofer: Morel MW168 - 88db, 4ohm
Tweeter: Vifa D26NC05 - 90db, 6ohm

So, if not resistors, then what components, or wiring needs to be done to get the system to a 4ohm nom. impedance?

By the way, I thought I'd try a car audio x-over first, because I dont have any test equipment, and most factors like group delay, etc. aren't as important.

I'm also confused about how to wire my 2 circuits. Your statement, again - told me I was wrong, but didnt advise the correct way to go about it.

You say doing my suggestion will yield an 8ohm system impedance. How do you get 8ohms from paralleling two 8ohm loads? You also say that will basically reverse my crossover (open the circuits), so how do I wire my positive and negative leads into my input terminals, if not parallel?
 
What Stu is saying is that because each of the 8ohm loads work over a different frequency range, only one would be visible to the amplifier at a time and so it would see an 8 ohm load. This is a seriously simplified statement but I hope it helps.

Anyway, 2am here. Time for sleep.
 
AudioFreak said:
What Stu is saying is that because each of the 8ohm loads work over a different frequency range, only one would be visible to the amplifier at a time and so it would see an 8 ohm load. This is a seriously simplified statement but I hope it helps.

Anyway, 2am here. Time for sleep.

Only one at a time? That would mean the music would have to be playing only above or below the x-over point at any given time.

Is this due to faulty wiring, or.... am I just totally lost?
 
Lusso5 said:


Only one at a time? That would mean the music would have to be playing only above or below the x-over point at any given time.

Is this due to faulty wiring, or.... am I just totally lost?
I hate to say it but you are totally lost. In an oversimplified description: the amp only sees the woofers impedence from the crossovers freq on down and the tweeters from there on up. There really is some overlap but below the xover freq the twetters appearant inpedence rises(thats why is gets less power below the xover freq) and above the woofers rises. It is not one at a time but one at any given frequency.
 
markp said:
I hate to say it but you are totally lost.

Sure you hate to say it..... Just kidding.

markp said:
In an oversimplified description: the amp only sees the woofers impedence from the crossovers freq on down and the tweeters from there on up. There really is some overlap but below the xover freq the twetters appearant inpedence rises(thats why is gets less power below the xover freq) and above the woofers rises. It is not one at a time but one at any given frequency.

Well sure, I realize this.

Thanks for the reply, but I really need something other than an "oversimplified" description to figure this out.

Again, thanks for the reply.

But what I can deduce is this:

If I build two x-over networks, say each with 8ohm speakers (no resistors in the networks - as all the primers schematics are), my 2-way system will have a approx. an 8ohm nom. impedance. So what we end up with is basically an average impedance?

So will my 4ohm woofer and 6ohm tweeter system end up with a 5ohm imp average? (My intuition would tell me this isnt right and the woofer would have to be given a heavier factor in the calc, since it is playing lower frequencies and pulling lower impedance numbers, but as you said "I'm totaly lost".)

How to hook them up? I'm still waiting on help with that, because apparently parallel wiring them into the amp will cause open-ended circuits. All the crossover primers show the x-over schematics as two separate schemcatics for tweeter and woofer, but dont say anything about special wiring, so I just assumed paralleling those leads into and amp would be fine.
 
You should parallel the xovers into the amp. There is no real average impedance in this situation. If you had two woofers on the same amp it would be averaged but not a woofer and tweeter as they operate in separate freq ranges. The amp 'sees' only the driver in that range. This means that at 100hz you have 4ohms and at 4000hz you have 6ohms.
 
I'm wondering how critical it is to try a get a 4 ohm nominal load. I'm sure you're dealing with SS amps and since the actual impedance is going to vary across the frequency range anyway, wouldn't something close that sounds good be ok? It's going to have to be judged that way with the absence of measuring equipment isn't it? Maybe I'm lost too. 🙄
 
OK, so I wasn't totaly lost - just missing a few pieces and making things more complex than they should be for me at this point.

I'm ready to build these, although one thing is still stuck in my mind - by building the crossovers from the guides/primers online, they will have no resistors (other than the zobel on the woofer). Every crossover I have ever bought, or seen a schematic for online, has a resistor in it somewhere.

I'll get it drawn up into a schematic and find someone to host it for me so I can post it here, and hopefully that will get input.

Thanks Markp, Audiofreak, and Stu.

wouldn't something close that sounds good be ok?

Yeah, that is fine with me. I was just inquiring about the nitty gritty, for my own knowledge about the subject.
 
Timn8ter said:
I'm wondering how critical it is to try a get a 4 ohm nominal load. I'm sure you're dealing with SS amps and since the actual impedance is going to vary across the frequency range anyway, wouldn't something close that sounds good be ok? It's going to have to be judged that way with the absence of measuring equipment isn't it? Maybe I'm lost too. 🙄
Not critical at all. Just keep it above about 3ohms minimum and you'll be fine.
 
Lusso5 -

There is no need to host the pictures yourself - just attach them with your post and images will appear as images. (I can host them for you too if they aren't too big)...

Sorry about not giving you a straight answer before - I'll try and give you some answers now - although some of them have already been explained.

If you are going to use resistors, I would only use them on the tweeter - what you need to use is an l-pad, as shown below (the l-pad is the closest two resistors to the tweeter).

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


The l-pad, as I mentioned before also serves the purpose of matching sensitivities - by my calculations, for a 2 dB pad, to give 4 ohms for a 6 ohm tweeter, you need a series resistor of 0.82 Ohms and a parallel resistor of 6.75 Ohms (just get the closest values possible, and it should be ok).

As for paralleling the inputs to the crossover, they don't parallel like resistors, because they aren't resistors - they are complex, frequency dependent loads - the maths to solve it might get a little bit tricky, but the high/low approximations are easy enough to show why they don't simply parallel.
 
Status
Not open for further replies.