Anyone know any good websites / forums for restoring Vintage amps?

[ecc83]

The problem is guitar amps don't play fair!

In a "hi fi " amp the static and drive voltages are pretty predictable and the amp would not be driven into heavy distortion. E.g. the venerable Mullard 5-10 uses 47R1/4W components.

Then a screen grid will act as an anode given 1/2 a chance if the anode voltage dips too low and it will in a guitar amp because of the "abuse"! In fact, disconnect an anode supply and you can kiss the resistors and probably the EL84's goodbye in very short order!

And lastly the BSH principle. Having over rated components not only ensures a long life (and proof against abuse!) but also a 5 watt R dissipating 1/2 a watt will not get as hot and therefore be kinder to components around it. This is why it is always a good idea to make op valve cathode Rs 10Ws if you can as they are almost always next to and cooking the bypass cap.

Dave.

[dogface]

Originally Posted by ecc83

The problem is guitar amps don't play fair!

In a "hi fi " amp the static and drive voltages are pretty predictable and the amp would not be driven into heavy distortion. E.g. the venerable Mullard 5-10 uses 47R1/4W components.

Then a screen grid will act as an anode given 1/2 a chance if the anode voltage dips too low and it will in a guitar amp because of the "abuse"! In fact, disconnect an anode supply and you can kiss the resistors and probably the EL84's goodbye in very short order!

And lastly the BSH principle. Having over rated components not only ensures a long life (and proof against abuse!) but also a 5 watt R dissipating 1/2 a watt will not get as hot and therefore be kinder to components around it. This is why it is always a good idea to make op valve cathode Rs 10Ws if you can as they are almost always next to and cooking the bypass cap.

Dave.

That's not quite true. 1 watt power dissipation is 1 watt of heat, whether it's coming from a 1 watt resistor or a 10 watt resistor. The 10 watt resistor will feel cooler to the touch because the heat is spread over a larger suface area, but it's still giving off just as much heat.
However, the anode voltage dipping below the screen grid voltage would explain the extra power dissipation in the screen grid resistor. I suppose then it would be wise to leave the screen grid without a capacitor to maintain its voltage and let it go up and down with the anode. Then again, if the screen grid simply tracks the anode, isn't it acting as a triode instead of a pentode?

[WelshBoyo]

Depending on where you are based, my amp tech used to work for JMI and spent a lot of time on the AC30s.

He is a great guy and very very knowledgable.

PM me for more details.

[ecc83]

"That's not quite true. 1 watt power dissipation is 1 watt of heat,"

I was prepared for this!
The temperature of a body rises as to the heat input* (W) and its THERMAL CAPACITY! Therefore a 10W R handling 1W will not get nearly so hot (i.e. dgrs C) as a 1W handling 1W. So the temperature in the immediate vicinity of sensitive components will not be so high

Lower operating temp=smiles all round!

* basic stuff for designing heat sinks.

Dave.

[ecc83]

Here's a very rough rule of thumb.
If you can bear the outside of your fingers on a surface more or less indefinately it is probably sub 40C.

If you can stand but 2 or 3 seconds it is over 50C, and if spit boils!!!

Dave.

[jpfamps]

Originally Posted by Parker

Thanks everyone. I have tested the bulldogs today. One is fine while the other one has a very low fizzy output. Any ideas? The wire from the negative side to the speaker is not in good condition so could this be the cause?

Sounds like it needs reconing.

[jpfamps]

Originally Posted by dogface

The screen grid resistor baffles me a bit. I had to change a screen grid resistor recently in an amp I made a while ago. Even a 1/4 watt 100 ohm resistor should be able to handle 50mA (much more than than should be going through the screen grid anyway) but mine was going crackly, and most designs seem to use quite a high powered resistor. Is there a good reason for this which I'm missing?
And some designs have the screen grid resistor bypassed with an electrolytic (to hold the voltage steady) and some just have the resistor un-bypassed. What difference does it make to the sound? Perhaps I should just do some experiments myself but I was wondering if anyone here could shed any light on the matter.

For maximum power you want to keep the screen voltage as stable as possible during the signal cycle (ie not dipping with increased signal) which is why a capacitor is sometimes connected directly to the screens. Remember, back in the day power output was v-v-v-very expensive so every effort was made to maximize power output.

However, I would DEFINITELY NOT recommend this connection in a guitar amp, as you now have the opportunity to deliver considerable current to the screen.

Whilst the plate of a valve is quite a large structure, the screen grid is relatively small and so can't dissipate as much power as the plate. The EL84 data sheet gives 2W maximum for the screen grid compared to 12W for the plate.

Screen grid current, as explained above, increases massively when the power valves are overdrive (ie in guitar amps), so a resistor is employed to limit the screen current. Watching the valve screens in many amps and you can see them glow cherry red when the amp is played hard!

If the amp is driven into clipping the screen resistor will have to withstand much more power dissipation than the static power dissipation, so for reliability a 5*W part makes sense, especially as these aren't going to break the bank.

Obviously the bigger the resistor value the better the current limiting, which is why 1*K is often recommended in AC30s. The downside of this is that the screen voltage dips on clipping, thus limiting power (similar to power supply sag due to a valve rectifier) and occurs in all amps even genuine class A amps (which of course the AC30 isn't).

Whether this is audible is a moot point.

[dogface]

Originally Posted by ecc83

"That's not quite true. 1 watt power dissipation is 1 watt of heat,"

I was prepared for this!
The temperature of a body rises as to the heat input* (W) and its THERMAL CAPACITY! Therefore a 10W R handling 1W will not get nearly so hot (i.e. dgrs C) as a 1W handling 1W. So the temperature in the immediate vicinity of sensitive components will not be so high

Lower operating temp=smiles all round!

* basic stuff for designing heat sinks.

Dave.

Still not quite true imho. 1 joule of energy will not raise the temperature of a large body as much as it would a small body, but there's still 1 joule dissipated. Similarly, 1 watt of power dissipation for limited time (say 10 seconds) will not raise the temperature of a 10 watt resistor as much as a 1 watt resistor, but there's still 1 watt being dissipated. In a guitar amp the cathode resistor will be dissipating heat for as long as the amp is switched on and will be dissipating the same amount of heat energy whether the resistor is big or small. The general air temperature inside the amp will be the same regardless of the power rating of the cathode resistor.
However, a smaller resistor will get hotter, true enough, and "the immediate vicinity" is what matters to electrolytics in the immediate vicinity. I like to have electrolytics out of the immediate vicinity by having them spaced away from hot resistors instead of snuggling up to them. I take your point about the immediate vicinity rather than the ambient air temperature inside the amp.
I think I'll do some experiments with the screen grid resistors in my amp to see what differences I can hear as well as measure.

[ecc83]

" The general air temperature inside the amp will be the same regardless of the power rating of the cathode resistor"

Nope, sorry.

The air cannot be heated except by the resistor. A big resistor will reach a lower surface temperature for a given power input and therefore the air around can only attain that temperature.

This is the principle of the heat sink. A transistor on its own cannot handle, say 20W because its junction temp' would rise to destruction. Energy is pumped into a slab of ally and some of that energy is "lost" raising the temperature of that metal. Thus the device stabilises at a lower, safe temperature. Even more energy would be lost if we made the heatsink out of copper.

Dave.

[dogface]

Originally Posted by ecc83

" The general air temperature inside the amp will be the same regardless of the power rating of the cathode resistor"

Nope, sorry.

The air cannot be heated except by the resistor. A big resistor will reach a lower surface temperature for a given power input and therefore the air around can only attain that temperature.

This is the principle of the heat sink. A transistor on its own cannot handle, say 20W because its junction temp' would rise to destruction. Energy is pumped into a slab of ally and some of that energy is "lost" raising the temperature of that metal. Thus the device stabilises at a lower, safe temperature. Even more energy would be lost if we made the heatsink out of copper.

Dave.

Sorry, but I can't let this go unchallenged. Heat is one thing, temperature is another. Heatsinks don't work by absorbing heat, they pass the heat to the surrounding air and convection takes that heat away. That's why heatsinks are the shape they are with fins and as much surface area as possible. If they simply absorbed heat they could be just a lump of metal. Of course, a lump of metal absorbing heat work work ok for a short time, but eventually it's capacity to absorb heat would be reached and it would be at the same temperature as the thing it was trying to cool (if there was no heat passed to the surrounding air). The term "heatsink" is a misnomer. It is a heat conduit.
Heat is a form of energy, measured in joules. Power is rate of energy production, measured in watts, which is the same as joules per second. 1 watt power dissipation is producing 1 joule of energy per second, regardless of the size of the resistor. In a closed space the air within that space will absorb 1 joule per second and its temperature would rise indefinitely unless that heat is conducted/convected away. In an AC30 the air inside is badly vented but it isn't completely insulated so the temperature will rise until it reaches equilibrium with as much heat being conducted/convected away as is being produced inside. The average ambient air temperature inside the amp will not depend on the size of the resistors used, it will depend on how much power is dissipated and how well it is ventilated.
The air temperature right next to a hot resistor can only ever be as high as the temperature of the resistor, true, but not all the air inside the amp is in direct contact with the resistor. There's convection within the amp.