How Much Power can Cranked 100 Watt Superlead Produce?

[FleshOnGear]

Hmmm, that's eight 12" speakers rated 30W each? I've seen (well, I will in this case assume that Johan isn't having us on) Segeborn push a dimed 100W Marshall head into a single Greenback for about 3 minutes. It didn't fail, but he thought he might be pushing it too far. Heat will be the main factor there. Celestion are notoriously conservative in their ratings. Assuming that Jimi had the cabs loaded with Greenbacks, I would initially find it hard to believe. That said, he played with feedback a lot, and that will load the circuit in a continous fashion. Sounds excessive to me, but not impossible.

Taking the 170W from Pete's reference, that's 21.25W per speaker assuming they're all equal. I'd reckon that a 30W or even a 25W rated Celestion would handle that without too much trouble - also for prolongued periods of time.

Unlike the EV SRO, Celestion simply state a power rating, which - one can only assume - is based on RMS.

Maybe Hendrix wasn’t simply cooking the voice coils, but actually causing over-excursions. I might be wrong, but I always thought of burning up a voice coil and actually damaging the physical suspension of the speaker to be two separate things.

 

[XTRXTR]

@PelliX I just noticed this: Roland GP-8 w/ FC-100, EV-5, MIDI

I still have mine as well. The delay is great. I don't use it any more but could. I bought it in 87 I think. Stupidly at the same time sold my original Ibanez tube screamer with the chiclet foot switch.

 

[Pete Farrington]

Celestion https://celestion.com/blog/power-handling-all-you-need-to-know/

I recall seeing Celestion info describing how the power handling rating was obtained, basically the thermal EIA426A procedure noted on this Eminence page https://eminence.com/blogs/blog/understanding-loudspeaker-power-ratings

 

[PelliX]

@PelliX I just noticed this: Roland GP-8 w/ FC-100, EV-5, MIDI

I actually have two GP-8's, two FC-100's and just the one EV-5 (which I never use much, honestly). The modulation effects are pretty good, too. I don't care much for the distortion, but the rest is decent. It's part of the classic Status Quo setup - had to have it. The GX-700 is the next worthy follow-up that BOSS/Roland produced and it has a noise gate, configurable sequence of effects, configurable loop, a *very* good tuner.

The GP-8's are a bit dated now and particularly if you haven't had yours on for a while, you might want to check the battery. If you're the tech type, I can also send you a dump of the patches I have configured...

There's no 'proper' software for it, AFAIK, but I wrote a Ctrlr interface.

 

[XTRXTR]

Yeah I haven't powered on that thing since I bought this:

This is old now but I like it better, reverb and delay pre and post line them up how I want them. I mainly use the reverb and delay maybe some chorus for glassy cleans. Sorry for steppin on the topic.

 

[PelliX]

Never tried one of those to be honest, but waaaay back I used a Digitech RP-70 which was brilliant for its humble price tag. Also recently picked up an old Digitech Studio Quad 4 for some vocal processing mainly. Good kit, generally.

 

[XTRXTR]

I actually have two GP-8's, two FC-100's and just the one EV-5 (which I never use much, honestly). The modulation effects are pretty good, too. I don't care much for the distortion, but the rest is decent. It's part of the classic Status Quo setup - had to have it. The GX-700 is the next worthy follow-up that BOSS/Roland produced and it has a noise gate, configurable sequence of effects, configurable loop, a *very* good tuner.

The GP-8's are a bit dated now and particularly if you haven't had yours on for a while, you might want to check the battery. If you're the tech type, I can also send you a dump of the patches I have configured...

There's no 'proper' software for it, AFAIK, but I wrote a Ctrlr interface.

That looks cool will it run on linux? I might pull out my GP8 just to see if it is running.

 

[XTRXTR]

Celestion https://celestion.com/blog/power-handling-all-you-need-to-know/

I recall seeing Celestion info describing how the power handling rating was obtained, basically the thermal EIA426A procedure noted on this Eminence page https://eminence.com/blogs/blog/understanding-loudspeaker-power-ratings

Thermal limit...what part of the speaker are they monitoring? No Offense intended I respect your information and effort. I am not even trying to argue.

The following is how I felt after reading it.

I read that eminence thing and it seems like a long winded way of saying "F*** off, we know what were doing." Without saying anything about what they are doing. Are they measuring the pressure over time in a specific volume of gases while they run pink noise and use the ideal gas law to determine temperature? -lol

 

[PelliX]

That looks cool will it run on linux? I might pull out my GP8 just to see if it is running.

Aye, it sure will. It was written and used on Linux as it happens. Download Ctrlr and off you go. I did some VERY nasty Lua code for the bank selection. It works fine, but as Lua has no concept proper math... I got creative with the decimal point.

Lemme see if I can send you a zip in a DM.

Thermal limit...what part of the speaker are they monitoring?

The coil is the only part of interest temperature-wise, really.

 

[Pete Farrington]

Thermal limit...what part of the speaker are they monitoring?

As I recall, it’s the magnet / frame.
After several hours suspended in free air being operated at rated power, the whole assembly structure will get darn hot and reach a thermal equilibrium. From that temperature they derive the likely coil temperature.
Try touching the speaker magnet of any open back combo after it’s been cranked for a while - ouch, be careful

 

[XTRXTR]

As I recall, it’s the magnet / frame.
After several hours suspended in free air being operated at rated power, the whole assembly structure will get darn hot and reach a thermal equilibrium. From that temperature they derive the likely coil temperature.
Try touching the speaker magnet of any open back combo after it’s been cranked for a while - ouch, be careful

Sounds reasonable, I suppose at the same time they could measure the coil impedance in a series connection and watch it over time until it is destroyed and then they would know the point just before melting. Compare the thermal data versus impedance data over time. A thermal limit could be averaged and tested against that point every so often for quality assurance. Measuring the audio output in decibels to that point could be how they determine the db output rating as well.

 

[PelliX]

Maybe Hendrix wasn’t simply cooking the voice coils, but actually causing over-excursions. I might be wrong, but I always thought of burning up a voice coil and actually damaging the physical suspension of the speaker to be two separate things.

They are indeed two different things. You're only really going to get that kind of issue at the low frequencies, I'd say. Of course, I have no exact knowledge of what Jimi was doing or in what exact setup (though @Maggot Brain could probably add some factoids), but I think the feedback causing coil burnup seems *more likely* to me.

 

[FleshOnGear]

They are indeed two different things. You're only really going to get that kind of issue at the low frequencies, I'd say. Of course, I have no exact knowledge of what Jimi was doing or in what exact setup (though @Maggot Brain could probably add some factoids), but I think the feedback causing coil burnup seems *more likely* to me.

Yeah, burning up the coils seems more likely. Though, I wonder if dive bombs through a Fuzz Face could deliver enough bass to literally *blow* a speaker.

I remember years ago I ordered an old 1990 8x10 cab through a place called Daddy’s Junky Music - sight unseen. When I got the cab, only 4 of the original 7442 speakers worked. Two of the broken ones had voice coils that were fused in place - there was no movement of the cone possible. The other two bad speakers looked like they had been hit with something from the backside. The cones were punched forward and awkwardly creased. It looked like someone had managed to burn up two, and blow two.

 

[playloud]

115W @ 4% distortion, 170W @ 10% distortion

From https://www.drtube.com/schematics/marshall/jcm800-mv-ld-manual.pdf

All that prevents a 100W rated amp from putting out a 200W square wave is power supply sag. ie as with any unregulated power supply, as current draw increases, voltage decreases. Less voltage to the valves means less power output.

What does "average distortion" mean here? Assuming THD, these numbers seem a bit odd, no?

Given Prms = Vrms^2 / R, I believe we can rewrite as Prms = (1+THD)^2 * V1rms^2 / R, where V1rms is the RMS voltage of the fundamental (1k). But then the supplied measurements would imply 170/115 = 1.1^2/1.04^2 ~ 1.12!

Am I missing something?

 

[Metroman]

Hmmm, that's eight 12" speakers rated 30W each? I've seen (well, I will in this case assume that Johan isn't having us on) Segeborn push a dimed 100W Marshall head into a single Greenback for about 3 minutes. It didn't fail, but he thought he might be pushing it too far. Heat will be the main factor there. Celestion are notoriously conservative in their ratings. Assuming that Jimi had the cabs loaded with Greenbacks, I would initially find it hard to believe. That said, he played with feedback a lot, and that will load the circuit in a continous fashion. Sounds excessive to me, but not impossible.

Taking the 170W from Pete's reference, that's 21.25W per speaker assuming they're all equal. I'd reckon that a 30W or even a 25W rated Celestion would handle that without too much trouble - also for prolongued periods of time.

Unlike the EV SRO, Celestion simply state a power rating, which - one can only assume - is based on RMS.

Its well known though, that he blew speakers up on a regular basis. Theres no arguing that. And when they became available he changed from the 25w M -75hz series, to the 30w H -55hz series Celestions.
And in the late 60s he had his USA touring amps modded to run 6550 tubes. Amps were modded by West Coast Organ and Amps.
Several years ago the guy that did the actual mods was on the Metro Amp forum, and told about what all the mods he did to Hendrix late 60s amps.

 

[Metroman]

Perhaps that 240W rating wasn’t based on a continuous signal power input?
See how an EV SRO rating decreases as the conditions become more arduous? https://www.manualslib.com/manual/43766/Electro-Voice-Sro-12.html

So is it a 60W, 150W or 300W speaker?
As I use mine with an overdriven valve guitar amp, I regard it as good for 50W, suitable for use with at most a cranked 35W amp (with a saggy valve rectified HT supply).

In Hendrix late 60s 100w Super Lead Cabinets, with the 100w LOGO. Those cabinets came with 30w 55HZ H Series speakers. 120w RMS per 4 x 12 cabinet. And he used 2 -4 x 12 cabinets. 240w RMS.

 

[TheKman76]

The coil is the only part of interest temperature-wise, really.

Not so, sir. Glues in the 60's were not what they are today. Under the wrong (right) circumstances separating the spider and/or the cone from the former was certainly possible. It only takes a little softening to cause thermal run away if the cone doesn't move enough.

Of course, if one coil manages to short as a result (thermal expansion, maybe rubbing in the magnet gap), the other in a series pair is seriously at risk. If a pair fails open there's a good possibility the remaining speakers will see higher power from a tube amp.

Marshall's closed back cabs only makes this potential worse.

 

[ThreeChordWonder]

Wattage is irrelevant except for component sizing. It's decibels that count.

 

[Pete Farrington]

In Hendrix late 60s 100w Super Lead Cabinets, with the 100w LOGO. Those cabinets came with 30w 55HZ H Series speakers. 120w RMS per 4 x 12 cabinet. And he used 2 -4 x 12 cabinets. 240w RMS.

Thanks! Based on the JCM800 1959 info, that sort if thing should be fairly unlikely

But as it seems to have happened often, a few further points come to mind -

1 if the output was typically 115W @ 4% and 170W @ 10% distortion, if pushed to a squarewave (which I think equates to 50% distortion) the output power will probably be a bit higher still.

2 it’s difficult to speculate what the output of late 60s 1959 was. Especially as I seem to recall Hendrix had 6550 installed in them (?). Didn’t some have really high HT supply voltages?

3 I recall reading that 60s Marshalls had mains transformers with a 110V setting for North American use, ie rather than 117 or 120. Connecting a 110V tap to 117 or 120V will boost up the secondary voltages, heater and HT, above their intended levels, which will increase the amp’s power output. Power follows the square of voltage, so a 9% increase in HT voltage potentially provides a 19% increase in power output.

4 Hendrix played a lot of festivals, which seem likely to have used on site generator provided mains electricity. Perhaps they weren’t always putting out a perfect 120V sinewave. Again, if not, that would affect the amp’s HT voltage and hence its power output.

 

[playloud]

I think we might be missing the "Occam's razor" explanation for Hendrix's blown speakers: user error. There is some backstage footage of Hendrix's shows floating around (e.g. from the Isle of Wight concert), and it looks fairly chaotic to say the least! It wouldn't be at all surprising if some heads had only one 4x12 connected, or instrument cables were used instead of speaker cables, or speaker cables were faulty etc. If that happened a few times across the hundreds of shows Jimi played in his sadly short career, then we have our "pattern".

It also seems the blown speakers were more common with the earlier 20/25W Celestions, before he switched over to the 30W ones (although I'm no expert on Hendrix folklore!)

I would also suggest, at the risk of being a little cynical, that blown speakers are good for the "rockstar mythos" and cases may have therefore been exaggerated

1 if the output was typically 115W @ 4% and 170W @ 10% distortion, if pushed to a squarewave (which I think equates to 50% distortion) the output power will probably be a bit higher still.

If a true square wave (or something close) was achieved, I would imagine RMS power is a secondary concern. Consider dV/dt at the transition point - good luck to the voice coil!

Of course, with its wild/assymetric clipping, I don't think a fuzz face is likely to get you anywhere close to a square wave output.

Edit: also not sure about this part. The amplitude of output voltages generated by the amp doesn't increase linearly with mains voltages into the amp, does it?

Connecting a 110V tap to 117 or 120V will boost up the secondary voltages, heater and HT, above their intended levels, which will increase the amp’s power output. Power follows the square of voltage, so a 9% increase in HT voltage potentially provides a 19% increase in power output.