power rating wakeup

[supah]

Okay.

I thought that if, for example, a Marshall is rated at 18 watts, (as the 18 watt is), then 18 watts is the maximum power you would expect the speaker to see.

However I was talking to Jim from Southbay Ampworks, the creator of the Scumbacks (who I would trust with this kind of stuff) and he said that that's more of the power rating you expect to see clean, with the volume at 4-5, and that he recommends doubling the power rating for the speaker.

So for an 18 watt marshall, use at least 36 W speaker.
For a JTM45/100 marshall (100W), use speaker configuration of at least 200W.

Here's a PDF on his site where he talks about this...
http://www.southbayampworks.com/sd/speakermounting.pdf

If this is true I feel lucky because I just happened to have a vintage 30 which is rated for 60W. But before I decide on speakers based on this I wanted to get some 18watt.com opinions.

[krx]

It depends how you treat the speaker. Doubling the power rating is extremely conservative. That's what they're telling you because they don't want to deal with people blowing up speakers.

[kleuck]

Ac30 uses two 15 watts speakers, and they don't blow-up so often (but it happens)
Keep in mind that if you push your amp's power stage deep into saturation, the speaker(s) doesn't really see 15 or 18 watts, but rather twice these numbers.
And sound-wise you can encounter issues, depending on the speaker before blowing it.
I have an amp rated for 25 watts, but which was actually and internally limited buy the schematic itself (bad design and preamp distortion etc).I put a JensenP10R (alnico) rated 25 watts in it, and i loved it (after a long long break-in)
Then i modified the amp to get the most power out of it (335 volts and EL84 PP so similar to a 18 watter) and the P10R just couldn't stand this extra power and became lose and undefined when pushed.

[zaphod_phil]

I thought that if, for example, a Marshall is rated at 18 watts, (as the 18 watt is), then 18 watts is the maximum power you would expect the speaker to see.

However I was talking to Jim from Southbay Ampworks, the creator of the Scumbacks (who I would trust with this kind of stuff) and he said that that's more of the power rating you expect to see clean, with the volume at 4-5, and that he recommends doubling the power rating for the speaker.

So for an 18 watt marshall, use at least 36 W speaker.
For a JTM45/100 marshall (100W), use speaker configuration of at least 200W.

Here's a PDF on his site where he talks about this...
http://www.southbayampworks.com/sd/speakermounting.pdf

If this is true I feel lucky because I just happened to have a vintage 30 which is rated for 60W. But before I decide on speakers based on this I wanted to get some 18watt.com opinions.

I have a lot of respect for Jim, but I don't really agree with what's being said here, and the facts actually prove otherwise. Amplifier power ratings aren't really much more than "marketing watts". Guitar loudspeakers are generally made to handle a lot more abuse than hi-fi speakers, and in fact many people prefer their tone when they're being pushed towards their limits. Marshall's original 1974 1X12 18W combo featured a 20W Greenback speaker. The recent hand-wired RI version also features a 20W Greenback RI. There is no history of these speakers getting destroyed or burnt out in these amps, although plenty of transformers in the RI 1974X have gotten burnt out. A lot of 18W clones have also been built with a single 15W rated Celestion Alnico Blue, and again there is no known history of these speakers failing in 18W amps.

Also BTW those 60W V30s have an upper end spike in their frequency response that tends to over-emphasize the natural brightness of EL84 tubes. As a result V30s aren't much liked in 18W amps by most people, although there are always exceptions to every rule.

Then i modified the amp to get the most power out of it (335 volts and EL84 PP so similar to a 18 watter) and the P10R just couldn't stand this extra power and became lose and undefined when pushed.

OK, you didn't like their response, but they didn't burn out or fall apart.

[supah]

Zaphod - my 18 watt uses 6v6 not EL84s, the vintage 30 is ok but I really wanna try the greenback variations and see how i like that.

And also like you mentioned the one thing I'm afraid of is TOO much headroom on the speaker, probably some breakup on the speaker is nicer. But I have to try to see.

So it seems that I'll be fine with a G12H30, 30W-type speaker, in which case I will replace my V30 with something along those lines now that I have your blessing.

For my jtm45/100W head you think i'll be fine with two of the 65W scums?

[zaphod_phil]

So it seems that I'll be fine with a G12H30, 30W-type speaker...

Yes, that will be plenty fine with an 18Wish amp. Loud and full too.

For my jtm45/100W head you think i'll be fine with two of the 65W scums?

Yes, that should be perfect. IIRC those amps put out somewhat less than 100W anyway. The opposite is the case with early 100W Plexis, which put out around 120W clean and up to 200W distorted.

[kleuck]

Then i modified the amp to get the most power out of it (335 volts and EL84 PP so similar to a 18 watter) and the P10R just couldn't stand this extra power and became lose and undefined when pushed.

OK, you didn't like their response, but they didn't burn out or fall apart.

Right (remember my english is far from perfection )
They are rated 25watts, but i can't imagine them sounding good in a 25 watter, i think their natural compression is at its best for a 10 to 15 watts amp.
I just wanted to point out that electrical power rating is not the only issue, especially as you said, when one wants to push not only the amp but also the speaker, it can be difficult to find just the right one, we can't buy and test all the speakers from the market
I was considering the Red Fang", but did not buy it for this reason (despites its 50 watts rating): bought a Lil'Buddy instead, wich is great and powerfull, but exhibits no compression and very little if any distortion in this amp.

[hotair]

Might not be too much help on the current dilemma as I am not familiar with the speakers but more just food for thought.

Made a trade 30 years ago and ended up with a couple alnico Celestion speakers. One guitarist ripped in them for a few months with a Fender Twin Reverb, 100w+, and he was never meek with the volume control.

They were fine, then our rhythm guitarist started using them because his speakers didn't cut it. Don't remember what amp it was other than it was rated for 150w. Eventually the speakers developed a buzz, oh well that is life. Still sounded good (and what is a bit of buzz with all the distortion going on?)

Eventually they lost too much power and we put them aside. Dissected them years later and found that the voice coils unraveled but did not burn up. Called up Celestion for some recone kits but they answered back with sorry none availible anymore and a spec sheet from 1967. Seems each speaker was rated for 5w rms and something like 15w peak. A little less than we fed them.

At 100w they were happy, 150w the glue holding the coils together let go. The voice coil forms got a bit banged up with the excursion.

Which brings me to how speakers get blown. Two much heat or too much excursion. Two many watts can overheat the insulation and cause the coils to melt together (shorts in between the coils). Little bits and pieces can fall off and mechanically jamb up the voice coil gap making a scratching sound. Because the speaker is not following the amplifier signal it actually absorbs more watts (won't get into it now, just trust me) which can cause it to overheat.

Mechanical damage can occur by too much power but in the form of low frequencies. An amplifier watt of high frequency energy may put more heat into a voice coil than low frequency energy but the lows do the mechanical damage. (Open back speakers get more stress than ones in a sealed cabinet as long as the box is not too big.)

Now as different people play differently, some hitting the lows more, some hitting the highs, so they would be putting different kinds of stress on the speaker with the same amp. So how is the manufacturer to decide how to rate the speaker? Sort of take a best guess.

I like the idea of the speaker being rated for more than the amp is rated for. Ar some point the amp will be cranked to the max and I would like to see the speaker survive (I am cheap and hate buying new speakers). As long as the speaker is efficient I like burning up less watts.

I built a small practice amp, 15w with an 8" speaker. Lent it to a guy to try out and he asked how much he could turn it up. I said all the way, To make sure it would not get damaged I plugged in my bass and over drove it. It is a survivor. Mind you I built it as a bass amp.


Now in terms of mechanical and thermal damage to speakers, I had a 12" that had a bit of a buzz and since it was not the highest of quality I decided to give it a chance to go out in style. It was an eight ohm speaker and I plugged it into a 120v wall socket. Fed it 1500w rms. Lasted all of five seconds. It had never made so much sound in its life. Sustained both mechanical and thermal damage. I think it drove itself out of the voice coil gap and then burnt up in the air.

Like they say on TV, don't do this at home folks.

[zaphod_phil]

I like the idea of the speaker being rated for more than the amp is rated for. Ar some point the amp will be cranked to the max and I would like to see the speaker survive (I am cheap and hate buying new speakers).

In the world of guitar amps and guitar speakers, it seems that the speaker ratings are excessively conservative. So we get 15W rated Celestion Alnico speakers running perfectly happily with an 18W amp, that may possibly putting out around 25 distorted watts when cranking, and sounding awesome at the same time. The idea of running them off a 100W F*nder twin reverb without mishap, however, really blows my mind. I think what's more important is whether you want some speaker distortion in your tone or not, and factor that into your decision. If you don't want speaker distortion then maybe buy some 100W rated JBLs.

I had a 12" that had a bit of a buzz and since it was not the highest of quality I decided to give it a chance to go out in style. It was an eight ohm speaker and I plugged it into a 120v wall socket. Fed it 1500w rms. Lasted all of five seconds. It had never made so much sound in its life. Sustained both mechanical and thermal damage. I think it drove itself out of the voice coil gap and then burnt up in the air. Like they say on TV, don't do this at home folks.

[Dana-L]

This tech note from Peavey includes a nice explanation of how a 100 Watt amp can blow a 200 Watt speaker:

http://www.peavey.com/support/technotes ... pter_2.pdf

In summary, pumping a square wave through a speaker is much more stressful than a sine wave. They suggest that the speakers power rating be about 2.5 times that of the amp.

Cheers,

-Dana

[zaphod_phil]

The theory is basically valid. However, that particular article is mainly talking within the context of PA amplification, rather than guitar amplification. And there's also a not so hidden plug for big PV speakers in there. The experience of many years of guitar amps and speakers show that you do not necessarily need speaker power rating to be 2.5 times that of the amp, due to various electro-acoustic effects not discussed in the article. Often 1 times works fine - sometimes even slightly less, as in the case of 15W rated Celestion G12 single Alnico speakers running in Marshall 18W amps. IMO theory always needs to be tested in the laboratory of real life.

[CurtissRobin]

If theoretical considerations were enforceable and enforced rock wouldn't exist. We'd go to extremes to avoid driving tubes off the linear portions of their response curves and everything would be clean, clean, clean*. Speaker cones and voice coils would last forever and Dennis Hopper would've had a really short movie career.

KennyO

*On a saxophone forum I saw this:
KennyG gets on an elevator and says, "Man! That rocks!"

[skeezix]

In PA context, the actual practice these days is the opposite -- you power the speakers with an amplifier 1.5 to 2x the speaker's rated power. Then you structure the gain of the system so you maintain headroom and never even approach sending the speakers a clipped signal. 'Course, you fail at that, you can toast some pretty pricey drivers.