Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum. New: view Recent Searches.
New: visit Schematic Hell!
The sunn still shines online!

Listen to great tunes streaming live right now!

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  
previous: lonzo If an amp has a 4 ohm spkr out and ... -- 1140168861 View Thread

Re: 4 ohm or 8 ohm - Difference in Sound?

2/17/2006 12:00 PM
Re: 4 ohm or 8 ohm - Difference in Sound?
*I mean, does an amp react differently because more of the transformer is used?*  
I don't see how it would be able to react exactly the same. Whether this one factor is the most significant would be difficult to know for sure.  
*Let's just say two identical cabinets are loaded with the same type speakers (your choice) except one is wired for four ohms and the other wired for eight ohms.....thoughts?*  
In a nutshell I think decent percentages of better players with nicer guitars and more demanding techniques will prefer one over the other. At least that's my results from limited performance of this exact experiment. As always there's no accounting for taste.  
A couple technical factors may be significant, tube amps have abysmal performance when it comes to damping factor. It is accepted that the steep winding ratios of the OPT due to the need to match such disparate impedances leave little ability to dampen or control any speaker load compared to an equal-powered solid-state amp. This is more important for bass and its no surprise that for most bass players when *tightness* is the most critical tone factor they rely almost exclusively on solid state power amps.  
If you were desperate to get the most load damping you could from a tube amp in the simplest way, you would probably want the OPT to have a 1:1 winding ratio, but that would mean you needed a speaker load in thousands of ohms. Things like this have rarely been done and are not considered mainstream or practical. Plus IIRC you still don't get half as much damping as you do from solid state.  
At the other extreme would be loads closer to zero ohms, of course at zero ohms its not good and theoretically as well as practically there is zero damping as expected. Any OPT matching ratio in betwen these two extremes will have damping performance some where in between also.  
Since real-world speakers dwell in the close-to-zero ohm range (especially the 4ohmers) we have always lived with little-to-nearly-no damping other than that provided by negative feedback (NFB) from the OPT, on amps where this circuitry is present and when they are not operated beyond the range where this feedback fails to continue compensation.  
I would say if you like cranking amps to maximum power (using more-than-adequate speakers of course) then you have usually exceeded the effective damping range of the NFB and are relying mainly on the damping provided by the tubes themselves as leveraged by the winding ratio.  
Big lever.  
What this means has to do with EMF or electromotive force, which is a good catagory for speaker energy to be in. I guess you could say that if speakers speak a language it is EMF. You send them electrical energy and they put out the physical force. But also when you put physical force on them, they put out some electrical energy of their own. Like for instance if you had your rig on stage but were not playing through it you would expect your speaker cones to be unmoving. Yup for the most part but if another player is cranking enough db from his nearby rig, you have probably never thought your cones are picking that up and moving a little bit in response but they are. As they move they put out an electrical signal through the same terminals you hook up the speaker wires to, and in this case your speaker is plugged in and everything. So this little signal goes up your speaker cord and into the 4, 8, or 16ohm tap of the OPT, then gets stepped up tremendously by its winding ratio and, less any effective NFB, appears as a much bigger signal on the plates of the power tubes. Where it doesn't do anything because you are not playing at the time anyway.  
But hopefully you get the idea that speakers are capable of behavior that not everyone recognizes.  
Along those lines since the cone is mounted as a spring-loaded motor, you can push on the cone by hand and let go and the cone will spring back into resting position, generating a little EMF in the process of springing back, and of course this electrical signal appears at the connection terminals. Or when anything else pushes on the cone, it springs back accordingly. Also when your amp pushes the cone it springs back only this time the amp is sending more power toward the cone than the cone is sending toward the amp. And this is happening continuously as you play. You could say power is capable of flowing both ways simultaneously but realistically the smaller signal just cancels a bit of the larger signal (except for those parts of the waveform which are neutral or reinforce the amp's reference output) plus the delay of the recoil is imposed in proportion to its relative significance.  
Dampening this recoil is one of the things that amps having better damping factors do better, in the absence of NFB it is simple math if you negelct the small coefficients; a ratio matching the tubes to an 8ohm load will dampen reverse EMF about twice as good as when the tubes are matched to a 4ohm load, this is with otherwise identical speaker assemblies except for the difference in voice coil ohms. This would be the more strict verbiage to represent the tight control of variables I assume is intended when it says *two identical cabinets are loaded with the same type speakers*. And a 16ohm matched load turns out to have a damping factor 4 times that of the matched 4ohm load. That's a factor of four. This can get pretty significant, depending on how you play you can more easily notice the difference between 4 & 16 ohms than you can between closer selections.  
You may judge best when you truly have two identical cabinets, not only to A/B them more easily in real time. But it can still be informative to experimentally rewire cabinets you already have. However, for instance with a traditional 16ohm Marshall 4x12 where you have four 16ohm cones, each pair of cones is series-connected to form a 32ohm load, then these two 32ohm loads are paralleled and the amp sees 16ohms. When the amp's signal gets there it drives two intimately (parallel) connected 32ohm loads and I think it can be seen how none of the individual cones can receive signal without it going through a different cone on the way there. Consequently none of the reverse EMF from any one cone can get back into the amp without it passing through an adjacent cone on the way either. As it is being reinforced by the reverse EMF from that adjacent cone at the same time. Even at its simplest it can be somewhat complicated.  
Rewire that cab so all the cones are in parallel and the amp sees 4ohms. Set the amp for 4ohm output and you've got the correct match again with the exact same cabinet & cones. Some things you might play can sound different if you listen carefully. This time your damping factor took a dive but now the amp can speak to each cone individually without being filtered by another cone in series. And the cones can speak back (their reverse EMF) more directly. But the 3 other low-imedance (well, equal impedance but the ohms are low) loads are in parallel with each cone so there is still filtration, this time by loading rather than series current flow.  
So maybe the most signficant factor at work is the difference in OPT ratio and resultant disparities in damping. Or maybe there could be differences in speaker performance more significant that dwarf incremental changes in damping. Some speakers are probably better at this than others, and consistentcy may count most, applying more so to multiple-cone cabinets.  
This would be in addition to any differences in coupling among different OPT taps, which seems like it could depend a lot on how they were wound.  
Probably let each person's ears decide.  

lonzo Thanks to each of you for respondin... -- 2/17/2006 2:39 PM