The Canadian Guitar Forum banner

1 - 14 of 14 Posts

·
Occasional CEO
JB Custom
Joined
·
1,921 Posts
Discussion Starter #1
How many watts does my cab need to be for my tube guitar amp?

We get questions like this all the time. So, I'm working on short videos like this with simple answers to get people started on their gear journey.


Naturally, as with all subject matter, you can go down quite the rabbit hole of debate. But, this advice will suffice for the greatest number of people asking themselves this type of question.

Let me know if there's any other topics or questions you'd like to see discussed.
 

·
Premium Member
Joined
·
17,846 Posts
Are tube watts different than solid state/digital watts (no)
 

·
Occasional CEO
JB Custom
Joined
·
1,921 Posts
Discussion Starter #4
Are tube watts different than solid state/digital watts (no)
Correct. Watts are Watts... But, a 100W tube amp will have different speaker power handling needs than a 100W solid state/digital power amp.

Excellent! Concise, informative and easy to understand and remember.

I would like to see videos interviews (similar to the one with @mhammer ) with @nonreverb and @zztomato. Both are longstanding and highly regarded forum members and local professionals in Ottawa.
We've got quite a few lined up for the next while, but I'll take a look!
 

·
Premium Member
Joined
·
25,584 Posts
A great start, Jon. I hope to see more on the same theme.

People should always consider their speaker cab to be a fuse in a box. Fuses are designed to burn up when more than some designated amount of current is passing through them, thereby blocking unsafe current levels from reaching the equipment and damaging it. They can be roughly divided into fast and slo-blo types, the latter requiring that the designated current level be pulled through for some period of time before they destruct, while the fast-blo type take no chances and self-destruct quickly, when the danger threshold is reached.

The wire in the voice-coil of most speakers is certainly of a more robust kind than is found in your average guitar pickup coil. But, it is a long way from the sturdiness of the cable running from the amp to the speaker terminals. And remember that the cone's ability to move back and forth quickly is not only a function of the cone, but also the mass of the voice-coil, so you can't make it too heavy by using super-thick wire.

As I have blathered on about in previous posts, there can be heat build-up in the voice-coil. If one wants to avoid that heat entirely, you simply use a large-enough gap between the magnet and voice-coil and employ a monster magnet. But the efficiency of the speaker is severely reduced when one does that. So, for a whole host of reasons, not the least of which is the sheer weight and shipping costs of the magnet required for such a large gap, one aims to design the speaker such that the gap between coil and magnet is very slender. That requires that the voice coil move back and forth in an absolutely flawless piston motion, with negligible "wiggle" even at highest volumes (where the coil is moving to its most extreme forward and rearward locations).

But let's go back to the fast-blo/slo-blo distinction for a moment. Fast-blo can't handle excessive current for more than a very brief moment, while slo-blo allow for a wee bit of heat build-up before blowing. In other words, slo-blo can handle infrequent supra-threshold blips in the current. In that respect, a speaker voice-coil is a bit like a slo-blo fuse. As long as the momentary burst of current passing through the voice coil does not generate too much heat, the coil won't blow. That will vary with the speaker's ability to not generate heat through friction, dissipate any heat that might occur, and how often those bursts of current take place (i.e., as little build-up as possible, whatever the source).

It is the nature of guitar signals that picking a string yields a brief peak before the string settles back down. That peak is not only much greater amplitude than the rest of the picked string's sustain, but also contains a lot of harmonic content. And remember that harmonic content makes the voice-coil ove back and forth more quickly. It is in the nature of our contemporary sonic preferences that we like all that harmonic content. Indeed,"distortion" IS added/exaggerated harmonic content that was not there in the original signal. Overdrive/distortion/fuzz attempts to extend the added harmonic content of the initial pick attack such that it continues for some period after that initial pick attack. If the voice-coil were some boyscout rubbing sticks together to start a fire, more prolonged faster rubbing of the sticks yields greater heat and suddenly there are enough sparks for a fire.

Which leads us to solid-state vs tube power. One of the things we tend to like tubes for is their often natural compression (i.e., holding a level somewhat steadier rather than having peaks and dips), and the ease with which we can produce desirable overdrive. In some respects, it is not the difference between tube and solid-state (whether analog or digital) that matters. After all, current is current is current, as far as your speakers are concerned. But the more typical use of tube amplification is to provide a near constant high output of harmonic content, which can result in greater heat buildup in the voice coil.

So with that overly long intro, I agree with Jon's suggestion of speakers requiring 1.5x the power-handling rating of the amp's output rating. In theory, one can use much lower-ratyed speakers with any amp; e.g., a 5W 3" speaker with a 200W bass amp. BUT, you have to be absolutely sure that the signal fed to the amp is sufficiently low amplitude that the amp's output will not/never exceed the limits of the speaker, or else you get this. (Notice that he plugs into a GIbson GA-5, a 5-6W rated amp?)

Series vs. parallel. Following the "fuse" train of thought, when speakers are in series, each speaker serves as a fuse for the others. So, if I have a pair of 4ohm speakers in series to provide an 8ohm load to the amp, their power handling is NOT the sum of their individual ratings, but is the rating of ONE of them. I recently built a cab with a pair of 4ohm 20W speakers, wired in series. The power-handling of the cab is not 40W. It is 20W.

Parallel is a slightly different matter. In parallel, the current from the amp is effectively "divided" between two or more paths, such that each path is seeing less than the total current from the amp's output. I recently put together another cab with 4 drivers. Each speaker is 8 ohms, so two in series is 16 ohms, and two 16 ohm sets in parallel brings us back to a total 8 ohm load. Those speakers are also rated at 20W power handling. Within each series pair, the power handling is 20W, just like the 4+4ohm pair described above. But, because two of these 16 ohm sets are in parallel, the total power-handling of the quartet is summed; 20+20W or 40W.
 

·
Premium Member
Joined
·
25,584 Posts
Excellent! Concise, informative and easy to understand and remember.

I would like to see videos interviews (similar to the one with @mhammer ) with @nonreverb and @zztomato. Both are longstanding and highly regarded forum members and local professionals in Ottawa.
Richard Vernon (nonreverb) has a thriving (I hope) business in the servicing of Hammond equipment but also sees his fair share of amps and often does backlines at Ottawa Bluesfest. I think he'd be an excellent interview.
 

·
Premium Member
Joined
·
12,847 Posts
Series vs. parallel. Following the "fuse" train of thought, when speakers are in series, each speaker serves as a fuse for the others. So, if I have a pair of 4ohm speakers in series to provide an 8ohm load to the amp, their power handling is NOT the sum of their individual ratings, but is the rating of ONE of them. I recently built a cab with a pair of 4ohm 20W speakers, wired in series. The power-handling of the cab is not 40W. It is 20W.

Parallel is a slightly different matter. In parallel, the current from the amp is effectively "divided" between two or more paths, such that each path is seeing less than the total current from the amp's output. I recently put together another cab with 4 drivers. Each speaker is 8 ohms, so two in series is 16 ohms, and two 16 ohm sets in parallel brings us back to a total 8 ohm load. Those speakers are also rated at 20W power handling. Within each series pair, the power handling is 20W, just like the 4+4ohm pair described above. But, because two of these 16 ohm sets are in parallel, the total power-handling of the quartet is summed; 20+20W or 40W.
Sorry, Mark, but the series power rating of a pair of drivers isn't quite correct. If you have two 20W drivers in series, they will safely sink (or absorb) 40W of total energy.

In parallel, both drivers see the same voltage across their terminals, but half the current through their voicecoils (it is split between the two). In a series connection, both drives see the same current through their voicecoil, but only half the voltage across their terminals, like a resistive voltage divider (KIrchoff's voltage and current laws).

Power is a product of voltage and current, so in both cases each driver is sinking half the power of the amp (assuming equal drive impedances, of course), so it is a safe load, from a power POV.


I agree with your comments about added harmonic components, something not considered for the most part in SS amps. And I enjoyed @jbealsmusic OP video --- well done, the internet can always use more concise, accurate information. Keep 'em coming.
 

·
Premium Member
Joined
·
25,584 Posts
How does a 20W speaker safely pass 40W of amplifier output such that it makes it to the second speaker? Certainly the added load of the 2nd speaker plays a role, insomuch as the load of the cab is now 8ohms rather than 4, so the amp is providing less overall output. In that respect, yes, the series arrangement will result in less effective wattage fed to the speaker, regardless of configuration. But, ignoring what the amp is essentially "permitted" to provide by varying the speaker load this way or that, if speaker-driver A can only handle X watts before heat dissipation capability is exceeded and it goes all fuse-ey on us, how can speaker-driver B provide any sort of insurance against that?

Consider the venerable TDA2030, used as the power chip in more "practice" amps than we can possibly imagine. The datasheet indicates that a typical maximum output power of the chip with a +/-14V supply is 14W, going into a 4 ohm load. Increase that load to 8 ohms and the amp's maximum clean output power goes from 14W down to 9W, 18 and 11W if we are willing to tolerate 10% distortion. So a pair of 4 ohm speakers in series draws less overall power from the amp, yes, but it does not change the 1st speaker's heat tolerance. It's not at all clear how the amp's current could be distributed in a way that stays under the threshold. After all, if I stick a pair of 1A fuses in series, and try to pull 2A through them, that first one WILL blow. Not trying to be snarky, or a belligerent contrarian, but something is simply not making sense. (I was going to say "something is not adding up", but maybe that would be snarky. :giggle: )
 

·
Premium Member
Joined
·
12,847 Posts
How does a 20W speaker safely pass 40W of amplifier output such that it makes it to the second speaker? Certainly the added load of the 2nd speaker plays a role, insomuch as the load of the cab is now 8ohms rather than 4, so the amp is providing less overall output.
You can't think of it as 'one speaker passing power to the next'. Audio power is AC and oscillates between +ve and -ve, it doesn't flow from one to the other. Which speaker is passing power to the other (do you follow electron theory or hole theory)?

You have to think to it as a system. Both speakers are connected across the amp's output. Both speaker will see the same current but each speaker will see half the voltage of the amp's total output. True, in this config, each speaker will act as a fuse (only a concern in tube amps, which you don't want to run into an open circuit). But the speakers equally share the power and neither is a conduit to the other, because that is just 'current' thought with no consideration to the voltages involved.
 

·
Registered
Joined
·
5,416 Posts
Concur with @High/Deaf , power rating in series adds.
It gets a little tricky if the impedances or individual wattages are unequal, but it is a function of the power fomulae (P=I x E) and can be calculated the same as resistors.

As a practical example, consider a 100W combo amp with two 12 inch speakers wired in series (ex., red knob twin). The speakers do not need to be 100W each, they share the power equally between them.
 

·
Premium Member
Joined
·
25,584 Posts
Interesting video. Glad he didn't go all the way, but then I wonder just what the speaker was sounding like as the smell got worse. Great tone up to that point though, and he's a pretty soulful player.

I may have been wring about some things in this thread, but the ability to dissipate heat buildup isn't one of them.
 

·
Premium Member
Joined
·
12,220 Posts
Which leads us to solid-state vs tube power
I didn't really see anything on SS regarding the 1.5 X formula. Is it the same as tube?
For instance, I used to have a Traynor Bloc 100 that had a 100w 4ohm speaker. Is that fine if the amp was maxed out?
 

·
Premium Member
Joined
·
12,847 Posts
might be fun video (Johan released it an hour ago):

going to watch it right now.

Thank you for the info Jon! Great video
Fun video.

Of note, he only exceeded the recommended power handling with the last two settings on the brake (-3 and 0, which are 50w and 100w respectively), although considering the amp was probably producing more than 100 watts, perhaps the -6 setting may also have been above the 25 watt spec of the speaker. These speaker ratings aren't a brick wall, as shown in this video. You can exceed that spec, even by quite a large margin, and the speaker doesn't immediately evaporate or anything. Good chance with a cleaner tone, we may have heard more speaker distortion, too.

I wonder, if after that thermal overload, the speaker was as good as before. I wish he would have given us an opinion on that.
 
1 - 14 of 14 Posts
Top