Load on valve amps

[Charles Jenkinson]

I've got two 8 ohm speakers which I need to wire up in a mono - stereo switching circuit for an XF212. Do I want the mono configuration - when both drivers are working, at 16 ohms or 4 ohms, in relation to typical valve heads one might use? I.e. I don't have a valve amp yet, am just trying to shoot with an elephant gun for starters. I do however have a small stereo solid state amp which is 4 ohm capable, but I think it makes sense to optimise the mono setting for potential valve amp power. Thank you.

[Bill Fitzmaurice]

The load depends on the amp tap rating. Unlike SS you can't go too low with speaker impedance, but you can go too high. You don't get more power with a lower impedance load, because what loads the tubes isn't the speakers, it's the output transformer.

[byacey]

While it's true catastrophic failure can occur from insufficient load on a tube output stage, It's not a good idea to go lower than the specified impedance either.

Tube output transformers are wound to drive a specific load impedance, while at the same time presenting a specific load impedance to the output tubes for maximum power transfer. Tubes by their nature control the current flow through the output transformer primary of a high voltage / low current power supply, typically anywhere from 300 to 500V while the current is in the 100s of mA range.

Any deviation from the specified speaker load impedance reflects on the impedance of the transformer primary. Here's an example of what happens: Let's say the speaker load should normally be 8 ohms, and the primary winding is designed to normally present a 4000 ohm load to the output tubes; With this load the amplifier is designed to provide 50W RMS. Mr. Guitar man thinking he'll hot rod his guitar amp, gets a hold of a 4 ohm speaker thinking it'll be louder. Once connected, the 4 ohm load causes the primary impedance presented to the tubes to drop to 2000 ohms, calling for twice the current through the output tubes. The tubes try and oblige, but because of their internal resistance, the plates inside the tubes run hotter because the voltage drop across them has doubled.

The power supply usually will sag because it's called to deliver twice the power it was designed for, and the majority of the extra power is dissipated across the output tubes as heat rather than being delivered to the load, as Mr. Guitar man would like. This situation is accompanied by earlier clipping. Running hard this way for any period of time, something will eventually burn up, usually the order of likelihood being the output tubes, the power supply fuse, or finally, the output transformer.

I've seen better quality amps that have taps on the output transformer that can be switched for the appropriate load impedance, enabling the user to select the load impedance desired.

As an exception, some amps like the famous Fender A673 Twin Reverb have an auxiliary speaker jack on the back panel; inserting a plug into this jack puts the second speaker in parallel with the internal speakers in the cabinet. This can lower the impedance to half (Fender recommends no lower than 4 Ohms), but because the output transformer is overrated for much more power than normal, and Leo Fender added parallel output tubes, the amp will safely withstand this extra load. However, most manufacturers don't build this extra "beef" into the amp, in the interest of maximizing profits.

This is the long winded answer, probably more info than what the average guy cares to know.

The short answer is, It's best to maintain the proper speaker load for what the amp was designed for.

[coolhandjjl]

Don't forget that valve amps designed for live music may be damaged if turned on with no speaker attached. Tube Hi-Fi amps less so (depending on how the amp is designed.)

[byacey]

That's because high power amps use larger transformers with more inductance and higher plate voltages compared to home hifi tube amps, although it can happen to smaller amps too.

What happens with no load on the output is the output transformer becomes a large boost inductor, with the same end result as an automotive ignition coil. Thousands of volts are generated when the transformer field collapses creating arcing inside the tubes, across the tube socket pins, and within the output transformer. $$$

[Bill Fitzmaurice]

While it's true catastrophic failure can occur from insufficient load on a tube output stage, It's not a good idea to go lower than the specified impedance either.

Tubes aren't bothered by a dead short. Many amps, including all Fenders, use a shorting jack on the output so that it won't be damaged if run with no load. On Fenders that's the Speaker out, and every Fender owner sooner or later found out that if you plugged into Extension Speaker by mistake the amp didn't work, as without anything plugged into the Speaker jack the switch remained closed, shorting the output. When I was young and stupid we ran Fender amps with 1 ohm loads all the time, with never a problem. You should run as close as possible to the tap rating, but if you do run other than the tap rating lower is OK, higher is not. On an 8 ohm tap the amp will be fine with 4 ohms, maybe even 2 ohms, but not with 16 ohms. All of this of course is exactly the opposite of SS, which always should be run at no less than the rated load impedance, and which isn't bothered at all if there's no speaker plugged in. The reason why is that with tubes current will always flow into the output transformer primary, and if there's no load on the secondary to accept current the transformer itself will take it all and fry. With SS if there's no speaker plugged in no current flows from the output devices. That's modern direct coupled SS, some very old SS from the 60s used output transformers just like tubes.

[byacey]

Users of Ampeg SVT amps should be aware that these amps are very sensitive to working beyond the specified load impedance. Because these amps are designed to work past the conservative design limits of the output tubes ie: extremely high plate voltages and low impedance plate loading, they will suffer extreme damage without a sufficient speaker load.

[byacey]

Tubes aren't bothered by a dead short. Many amps, including all Fenders, use a shorting jack on the output so that it won't be damaged if run with no load.

Idling with out signal or very low levels of signal will do no harm with a shorted secondary, but if you run high signal levels into a dead short for any extended period of time you will be replacing parts. The shorting speaker jack is the lesser of the two evils, compared to running the amp into an open load, where it will suffer immediate destruction.

As I mentioned, the old Fenders were a very robust amp and are somewhat of an exception, but the output tubes will still suffer over time working into too low of a speaker load. This is analogous to running too low a load on a solid state amp.

These amps have become collectibles, so I wouldn't recommend anyone testing the theory. Tubes can be had, but the original output transformers have become scarce.

Edit: I made mention of selectable taps on some transformers: Maximum clean audio power through the transformer occurs when the optimum plate load impedance by design is presented to the tubes; this plate impedance is a direct function of the load impedance. This is why manufacturers provide these selectable taps.

[byacey]

For those that wish to read in to the fine details of this discussion:
http://www.google.ca/url?sa=t&rct=j&q=& ... WM&cad=rja

Page 623 of the pdf sums up output load effects.

Edit: Chapter 21 offers additional information on the load /transformer /output tubes interaction.

[coolhandjjl]

I have a Peavey Classic 120/120 two channel tube power amp. On the Peavey users forum, it was mentioned that these use fly back diode circuitry to prevent disasters if a speaker became disconnected during use. I can definitely see the logic behind using the robust speakon connector.

[byacey]

I'm not sure what flyback diode protection is, probably some back to back zeners or a MOV clamp across the primary transformer winding. These would have to be selected to clamp above the maximum audio swing so as not to interfere with the normal operation of the amp.

A Speakon isn't really necessary in that application; Switchcraft 12A series jacks have proven more than reliable over the long run. Better yet is hard wired speaker leads soldered to the terminals of the speaker(s). That way somebody who doesn't know any better can't disconnect the load; Of course this will only work with a one piece cabinet.

[Mr. foxen]

Running valve amps into dead short if fine, as many people who have nothing plugged in but crank the volume anyway find out (but usually think they got lucky). Sometimes they blow a fuse anyway from being cranked right up. Not generally the transformer that frys though, usually there is a flashover somewhere, usually a valve base, sometimes on the transformer if the insulation isn't up to much or something is missing insulation/dirty, and that blows the HT fuse before serious damage. If someone then foil wraps the fuse, then the transformers cook.

[byacey]

Running into a dead short will not cause arcing because the output transformer is loaded, and as a result no back EMF created when the field collapses. Running into a dead short will cause excessive plate current due to the tubes trying to drive a zero ohm primary impedance, and excessive current through the output transformer primary. No back EMF is created to cause arcing.

Driving an unloaded output transformer will cause arcing. It's due the back EMF created by the magnetic field of the core collapsing and re-inducing very high voltages back into the primary windings. Exactly like the way an automotive ignition coil works.

This argument is getting old. I'm only posting this info to perhaps save someone from potential expensive repairs and headaches. I do have professional credentials and years of experience to back up the advice I'm offering here. Ultimately it's up to the end user to heed or ignore this advice; it's their amp, not mine.

[escapemcp]

This argument is getting old.

I've never used a valve amp, nor ever think I will need to, but this thread is interesting (for me). I never quite understood how valve amps worked (still don't 100%) but a few more bits of the puzzle have been revealed. I'm off to Google "How valve amps work" now - certainly better than doing any work today.

[Mr. foxen]

This argument is getting old.

I've never used a valve amp, nor ever think I will need to, but this thread is interesting (for me). I never quite understood how valve amps worked (still don't 100%) but a few more bits of the puzzle have been revealed. I'm off to Google "How valve amps work" now - certainly better than doing any work today.

Important bit to remember is all of the appeal of valve amp to guitar players is the ways in which they fail to work and become non-linear. The various ways that guitarists force them into not working properly and the discrepancies between that and what they think they are doing that 'improves their tone' [covers their playing], means a great deal of wrong info is on the internet about them. Also I'd guess about 5% of the actually good valve amp tech guys who've been scratch designing them since before they were obsolete were actually use the internet on more than a basic level, Chambonino.com is probably the major resource for old amps from someone who proper knows.