8ohm / 16ohm - impact

darcym

when picking speakers for say 2 x 2 cabs, how do you personally select an 8 or 16 ohm setup, do the speakers matter in terms of brand x model y likes 16 ohm better than 8, or in general do you find 8 is less harsh than 16 etc etc, what is the selection process for you, what matters to you.

digitalscream

I generally don't care. Essentially, there will be very little difference in the sound of the speakers themselves; the main influence is probably going to be the fact that you're using all of the windings on the output transformer for 16R as opposed to half for 8R.

And, critically, it'll make far less difference to you/the audience than standing a few inches/feet to either side (respectively).

ecc83

I would say there would be more difference in the cones that any due to the impedance but since you cannot swap coils, how could you know?

BTW , pedantic I know but an 8 Ohm winding is 0.707 of a 16 Ohms, not 0.5.

Dave.

ICBM

There are differences between the sounds of 8 and 16-ohm versions of the same speaker, but it's hard to generalise because it's outweighed by other factors such as series vs parallel wiring and what impedance taps the amp has.

I would try to optimise the cab for the amp if possible. For example if you have a Fender head you usually need a 4-ohm cab, so two 8-ohm speakers in parallel. If you have an Orange head you ideally want a 16-ohm cab so you can run two cabs if needed - the amp doesn't have a 4-ohm output - so two 8-ohm speakers in series. If you have a Mesa head with no 16-ohm output you want an 8-ohm cab so two 16s in parallel... etc etc.

If the amp is like a typical Marshall head with all the options then as a general rule running at 16 ohms sounds brighter and more lively, running at 4 ohms sounds deeper and thicker, and 8 ohms is somewhere in the middle - if all other factors are roughly the same.

mart

What is the difference internally between an 8ohm speaker and a 16ohm one? Is it the same number of coils of wire but a different gauge? Or different strength/type/size of magnet as well?

ICBM

The magnets are identical, the difference is in the gauge and number of turns in the voice coil.

ecc83

ICBM said:

The magnets are identical, the difference is in the gauge and number of turns in the voice coil.

Yes, an attempt will be made to keep the "Ampere Turns" constant* but also keeping the total cone mass the same.

*Term(probabaly achaic?)  for the force in a magnetic 'motor'. Obviously the greater the current, the greater the force but the fewer the turns the lower the force but more turns means more resistance so the two factors must go hand in hand.

Dave.

darcym

ICBM said:

I would try to optimise the cab for the amp if possible. For example if you have a Fender head you usually need a 4-ohm cab, so two 8-ohm speakers in parallel. If you have an Orange head you ideally want a 16-ohm cab so you can run two cabs if needed - the amp doesn't have a 4-ohm output - so two 8-ohm speakers in series. If you have a Mesa head with no 16-ohm output you want an 8-ohm cab so two 16s in parallel... etc etc.

I think this is probably the key bit of information I needed for how to approach this in the real world, thank you.

ICBM

ecc83 said:

Yes, an attempt will be made to keep the "Ampere Turns" constant* but also keeping the total cone mass the same.

*Term(probabaly achaic?)  for the force in a magnetic 'motor'. Obviously the greater the current, the greater the force but the fewer the turns the lower the force but more turns means more resistance so the two factors must go hand in hand.

Off the top of my head and without doing the maths, I *think* that 1.4 (sqr rt 2) times the number of turns of wire of 0.7 (1/sqr rt 2) of the cross-sectional area will give double the impedance and the same coil mass... but I could be wrong and it may be more complicated!

darcym said:

I think this is probably the key bit of information I needed for how to approach this in the real world, thank you.

No problem . It often does boil down to practical considerations rather than theoretical when there are many different factors like this.

I also forgot to say - if there's no obvious reason to prefer one over the other because the amp has all the options, then:

16 ohms has the advantages that it uses all the windings of the transformer - this is sometimes thought to be more desirable - and can be rewired to 4 ohms if necessary without changing the speakers.

8 ohms has the advantage that it's pretty universally compatible with any amp you may buy or use in the future, but *can't* be rewired to either 4 or 16 ohms without changing speakers.

darcym

This is where it call gets a bit confusing for me, so if you can educate me further on this, that would be great.

so
a.) why is 16 ohms more desirable when it uses all the transformer windings IF there is no real sounding difference
b.) I get your point about being able to reuse the same speakers in the different configuration, but 2 x 8 ohm speakers can give 16 ohm in series, (and I can see how changing the wiring changes them to parallel would give you 4 ohm, but how would you get 16 ohm in a 2 x 12 using 16 ohm speakers ? - or have I miss-understood what you said when you said 16 ohms has it's advantage ?

ICBM

darcym said:

a.) why is 16 ohms more desirable when it uses all the transformer windings IF there is no real sounding difference
b.) I get your point about being able to reuse the same speakers in the different configuration, but 2 x 8 ohm speakers can give 16 ohm in series, (and I can see how changing the wiring changes them to parallel would give you 4 ohm, but how would you get 16 ohm in a 2 x 12 using 16 ohm speakers ? - or have I miss-understood what you said when you said 16 ohms has it's advantage ?

a - slightly more efficient coupling, so either better power transfer or less stress on the transformer, or both. There is a sound difference too, it's just usually outweighed by other factors.

b - the advantage is for a 16-ohm cabinet, ie with two 8-ohm speakers in series. You can rewire them in parallel to give a 4-ohm cabinet, or even fit it with a switch. An 8-ohm cab (two 16s in parallel) can't be rewired - other than to 32 ohms, which is not likely to be useful.

lukedlb

What about the safety benefit of parallel over series? Two 16s in parallel produces an 8-ohm cab; should one of these speakers break or connection fail, the transformer won't have any problem seeing a 16-ohm (one 16 ohm speaker) cab. Whereas, the opposite could be fatal for a transformer, where two 8s in series produces a 16 ohm cab; should one of these speakers break or connection fail, the transformer will see an 8 ohm (one 8-ohm speaker) cab and start to cook.

ICBM

lukedlb said:

What about the safety benefit of parallel over series? Two 16s in parallel produces an 8-ohm cab; should one of these speakers break or connection fail, the transformer won't have any problem seeing a 16-ohm (one 16 ohm speaker) cab. Whereas, the opposite could be fatal for a transformer, where two 8s in series produces a 16 ohm cab; should one of these speakers break or connection fail, the transformer will see an 8 ohm (one 8-ohm speaker) cab and start to cook.

There is no real safety benefit.

If the speakers are in parallel and one fails, the remaining one will then have to handle nearly all the power of the amp - the drop due to a mismatch isn't that high, usually only about 20% - and will probably blow as well, so you'll just end up with two dead speakers and no load on the amp, instead of one dead speaker and no load on the amp.

It's best to make sure that the cab will properly handle the power of the amp, so there is then very little risk no matter how it's wired. It's actually fairly rare for valve amps to blow speakers as long as there's a reasonable safety margin.

lukedlb

ICBM said:

lukedlb said:

What about the safety benefit of parallel over series? Two 16s in parallel produces an 8-ohm cab; should one of these speakers break or connection fail, the transformer won't have any problem seeing a 16-ohm (one 16 ohm speaker) cab. Whereas, the opposite could be fatal for a transformer, where two 8s in series produces a 16 ohm cab; should one of these speakers break or connection fail, the transformer will see an 8 ohm (one 8-ohm speaker) cab and start to cook.

There is no real safety benefit.

If the speakers are in parallel and one fails, the remaining one will then have to handle nearly all the power of the amp - the drop due to a mismatch isn't that high, usually only about 20% - and will probably blow as well, so you'll just end up with two dead speakers and no load on the amp, instead of one dead speaker and no load on the amp.

It's best to make sure that the cab will properly handle the power of the amp, so there is then very little risk no matter how it's wired. It's actually fairly rare for valve amps to blow speakers as long as there's a reasonable safety margin.

Wouldn’t that suggest that a 50 watt amp set to 16 ohms using a single 65 watt (or higher) 8 ohms amp is fine? I thought this was a big no-no. Wouldn’t the previous example I offered produce this result?

ICBM

A 50W amp at 16 ohms into a 65W 8 ohm speaker would usually be fine. Not very good for the power valves, but no worse than that.

It would be the other way round if one speaker in a parallel cab blew though - 8-ohm amp into 16-ohm cab, which was an 8-ohm cab with both speakers working - but the same applies. The danger is a series cab - if one speaker blows you have an open circuit.

But unless you had two completely different speakers in the cab, a 50W amp into *two* 65W speakers is not going to blow either of them so it won’t arise anyway no matter which way the cab is wired.

If you had two 30W speakers with the same amp, you should be able to see the problem...

This is also why I don’t think that mismatching the power ratings of speakers in the same cab too much is a very good idea - you can end up with a cab rating which is actually lower than one of the speakers. eg if you have a 25W and a 75W speaker, the cab rating is 50W not 100W. So if they were in parallel, a 60W amp could blow the 25W speaker and then there would be a 75W cab.

That’s the only way parallel wiring is safer, as long as you don’t mind sacrificing the lower rated speaker.

lukedlb

Thank you @ICBM. Speaker selection and fears have long been a grey area I have risked nonetheless. Ohms, power rating and sensitivity can be confusing especially in larger cabinets as well as mixed speakers. 
Only this weekend I discovered that a ceramic speaker rather than alnico is a better solution for my redplate (fender sound) by providing a darker sound where I can increase the treble eq on the amp to sound more fenderish without any of the high shrill notes coming from a brighter and more open alnico speaker.

lukedlb

Returning to the op, why do marshall amps prefer 16 ohms and fender amps 4 ohms?

ICBM

lukedlb said:

Returning to the op, why do marshall amps prefer 16 ohms and fender amps 4 ohms?

For arbitrary historical reasons largely, in something like the way the US had 110V mains and we had 240V. Someone in the early days picked a particular impedance and then everyone else followed - but obviously, differently on each side of the Atlantic.

This actually has a great bearing on the sound of Fender and Marshall amps. The first Marshall JTM45 was a dead-on copy of the Fender 5F6-A tweed Bassman - which actually runs at 2 ohms - but Marshall used what off-the-shelf parts they could find in the UK, which for the output transformers were 16 ohm types. But since they didn't adjust the circuit to compensate for the higher transformer winding, the JTM45 ended up with substantially more negative feedback than the Bassman, giving a much tighter sound and a crunchier, harder distortion. The difference between four 10" Jensens and four 12" Celestions is also important of course...