Syntrillium Backup Forums :: View topic - How can you wire 2 -8 0hm speakers to 8 ohm amp??

sparky184 · Posted - Sun Jun 29, 2003 9:50 am

I have a peavey TKO 80 Puts out 80 watts at 8 ohms. it has a 15inch 8 0hm scorpion in the encloser. I got a swr workmans 15 inch encloser now and I would like to wire this all togehter but how can this be done to keep 8 ohms , or do I have to paralle the speakers and would 16 ohms be safer then running 4 ohms. This is a solid state amp.?????

GothicV · Location: Canada

unfortuantely, there is no way to wire it. I think it's best to wire it to 16 ohms and go from there. Luckily, solid state amps are not nearly as picky as a valve amp. You can run most types of solid state amps with a speaker load with no problems...

SteveG · Location: United Kingdom

Alternatively, you could check out 'will my amp work on 4 ohm speakers?', where you will probably find a satisfactory answer...

SteveG · Location: United Kingdom

You have to bear in mind that speaker drive units don't just have 'an' impedance. The impedance varies at different frequencies, and will reach its highest value at the drive unit resonant point - and if you have a reflex cabinet, the impedance curve will be somewhat different from an infinite baffle, for instance. And if, as in your speaker system, you have other speakers connected in parallel with the bass unit, only with impedance modifiers in series with them (that's what the capacitors are doing), then you are going to get a pretty complex impedance presented to the amp. If you measure its DC resistance, it may come out somewhere near 8 ohms, but in reality, at different frequencies, it will be all over the place. And your resistance measurement will only be that of the bass unit voice coil...

This is actually the reason that you shouldn't put speakers in series as a rule. That, and the fact that the volumes don't tend to balance out too well, and the damping factor gets completely wrecked - which finishes off any chance of bass control completely!

Transistor amps only have a nominal impedance rating to stop you blowing them up when you turn the volume up too far with a low impedance load. The more speakers you have, the more power the amp has to deliver. The amp doesn't actually care too much about this until the volume is turned up, and the poor output stage is then asked to deliver what amounts to far too much power into the load. That's when the transistors go into what's technically a second-order breakdown - and go phut! This doesn't tend to happen with FET output stages - they are inherently stable in exactly the way that ordinary bipolar transistor ones aren't! Basically, this means that when they are overdriven in this way, they tend to turn 'off', rather than trying to turn further 'on'.

So if you don't turn the volume up too far, then it almost doesn't matter how many speakers you put in parallel until you get down to a small fraction of an ohm. But don't try this - it really doesn't take very much to blow the power transistors out with 15 drive units in parallel across the output! On the other hand, most decent amps will cope pretty well with two pairs at once, as long as you don't abuse the volume control. There's a pretty good chance that if the sound doesn't sound distorted, then you are not going to blow up the amp. If it does sound distorted, then it probably means that even if the volume doesn't seem particularly loud, you are probably overdoing it.

With valve amps, the situation is completely different, because you are looking at the maximum power transfer rules, which demand matched impedances. This is because there's a transformer involved, which has a finite secondary impedance, unlike the transistor/FET transformerless amps, that don't have output transformers, and don't have to rely on matched impedances.

I suppose you can sum it up by saying that a transistor amp has a certain amount of power that it can deliver, and it doesn't really matter where it's delivered to, as long as the limit isn't exceeded. A valve amp may have the same amount of power, but it can only deliver this optimally into a matched load. The proof of this is a little surprising, but you don't need to know how it goes - you just have to remember that it has to be matched impedances if a transformer is involved, and it's not so important if there isn't one.

alofoz · Location: Australia

As always a very complete answer from Steve.

I'll just add that in some cases, especially in older transistor designs, feeding into 4 ohms instead of 8 ohms will result in the loss of the bottom octave. I'm referring to the amplifier response here, not the speaker response. As this should extend well below the speaker response it should not be too much of an issue.

I'm reminded of the fact that I started working at EMI shortly after Neville Thiele moved on from there. I still have an original copy of the "Proceedings I.R.E.E. Australia" where he first published his paper on vented speaker enclosures. The maths in papers like this showing the behaviour of speaker impedance in (and out of) the enclosures makes very interesting reading.

The results of Thiele's work live on in the Thiele/Small parameters often quoted for loudspeakers.

SteveG · Location: United Kingdom

jethrobodjean · Posted - Sun Jul 20, 2003 3:34 pm

jethrobodjean · Posted - Sun Jul 20, 2003 3:36 pm

SoloTune · Location: USA