Why is the Express rated at 36 watts?

[leadfootdriver]

Discuss.

[rdjones]

"Only" 400 volts or less (in most cases) on the plates, output primary impedance on the high side and being conservatively rated.

reddog

[geetarpicker]

Typically one might expect 50 watts or so off a pair of EL34s in push pull. However with the low B+ and smaller than average OT (with a higher than average primary impedance) that lowers the output quite a bit right there. Ken liked the tone of EL34s with lower than typical B+, plus it made the circuit more easier on output tubes even allowing 6V6 tubes among many other types to be used.

[M Fowler]

Amps such as the Orange Retro 50 do the same thing run 290-0-290 PT with a 60w rated 4k OT but can switch between 30w class A and 50w class AB.

The amp gets a ton of distortion at lower volume and you can really hear the tonal changes between what they call Class A verses Class AB.

The Express is the same way while we generally consider 300-0-300 PT some say the actual voltage is much lower on an Express so that makes me want to try a 245-0-245 Edcor matched withe 6K6 OT.

Mark

[JamesHealey]

I made an express with that 240 edcor transformer you speak of RJ sent it by accident instead of a 300-0-300..

Used it with a Pacific 6k6.

Never said anything but it sure as hell sounded fine to me!

[M Fowler]

James I think I will try that run the voltage lower in an Express to see how I like it.

[leadfootdriver]

I plan to use the cathode current method to bias. The data sheet for a the EL34 I plan to use gives a max dissipation of 25 watts.

Do I still use 25 watts in the math formula when calculating the bias?

[rdjones]

I plan to use the cathode current method to bias. The data sheet for a the EL34 I plan to use gives a max dissipation of 25 watts.

Do I still use 25 watts in the math formula when calculating the bias?

The data sheet rating is the physical limitation of the plate's construction and does not take the rest of the circuitry into account.
The bias point is a percentage of that rating depending on the class of operation, typically 70% or less in Class AB1.

rd

[leadfootdriver]

K, I did the math, and 2-EL34's biased at 70%=35 watts.

[jaysg]

Has anyone actually measured theirs? I had a stock Liverpool for a while. Depending on tubes, I got 13 to 26W clean.

[RevD]

Not sure how Ken rated it, for example a 100 watt marshall is like 100 watts clean, a good running 100 watter with tubes a burning into distortion can probably put out 150 watts, least ways that's what I've been told and I tend to believe it standing in front of one...

Regards,

Don

[RJ Guitars]

I made an express with that 240 edcor transformer you speak of RJ sent it by accident instead of a 300-0-300..

Used it with a Pacific 6k6.

Never said anything but it sure as hell sounded fine to me!

Slightly embarrassed and totally apologize while pleased that it worked out. Probably some other guy somewhere with a screaming hot Liverpool cursing me right now.... I have started peeling the label off the bottom of the Edcor transformers and putting on the top so I can know what transformer that was that I just mounted to that chassis. My short term memory sucks...

what was I saying?

Anyway, if you have ever done plate voltage measurements as you dial down an amp with a power scaling option, you'll notice that your ears strain to tell the difference until the voltage drop starts to get substantial. It would be a really cool piece of info to chart the operating point of the tubes as you dial the voltage down and see where they are at half volume, quarter volume, etc.... but that is a digression of where I was going - The point of this is that you can lose a lot of watts before you start to hear big differences and your tubes are enjoying an easy life... this does not adhere to the old Gerald Weber thought that tube amps sound best at the point they are about to blow up... YMMV

rj

[Cliff Schecht]

When I'm building a clone and going for a particular sound or trying to get the sound very close to an original to study the circuit, I put quite a bit of effort into getting the voltages close to the design specs. IMO this turns out to be critical to getting "that" sound. Part of "that" sound is getting the power tube plate voltage, transformer impedance and bias conditions right. The power scaling methods keep this rule in check because they scale both plate voltage and bias voltages proportionately (especially for fixed bias).

For cathode biased amps with too much or too little B+ I will sometimes swap around the output stage bias resistor until I get the voltage drop across the power tubes to be close to what the schematic specifies. This means the power tubes are working as the designer intended. But I prefer to use Zener's and power transistors to get the B+ down before the power tubes ever see anything.

A lot of people fuss over getting bias currents to a certain number or plate dissipation to a specific value (say 70%). If I'm building a clone of a circuit where I know the designer(s) knew what they were doing then I won't do this. Instead I will put the amp under the recommended bias conditions where I find the amps sound best anyways. My Express has always been at -30V and has never needed any adjustment. I put in whatever EL34's or 6V6's I have laying around and it sounds good every time (granted I'm usually using good tubes ).

In my Fender clones I quit putting any sort of adjustable bias, instead I will tack in resistors until the bias voltage reads what it is supposed to. In push-pull amps I will check the bias conditions to make sure that there isn't too much imbalance (too much causes excessive hum but common mode rejection decreases dramatically with even slight mismatches!). Because the tubes I use are usually NOS or vintage anyways, I don't have a dozen pairs to sort through and match. I do what people have been doing for 70 years, I buy two power tubes and put them in the amp!

In single-ended amps I have become lazy. The last one I built (Vibrochamp) still hasn't had a single voltage measured in it while operating. I built it, checked the wiring and e-caps with a function gen/DVM, put in tubes and flipped the power switch. It's been going strong for about a week now and sounds groovin every time I turn it on. I used the stock bias resistor (as I always do on Fender cathode biased amps) and the amp sounds great right out of the box..

[2tone]

Ken had told me it was rated at clean before distortion....36 watts

[RJ Guitars]

... The power scaling methods keep this rule in check because they scale both plate voltage and bias voltages proportionately (especially for fixed bias).

For cathode biased amps with too much or too little B+ I will sometimes swap around the output stage bias resistor until I get the voltage drop across the power tubes to be close to what the schematic specifies. This means the power tubes are working as the designer intended.

Using power scaling on my Rocket, about the time I get the volume down 50% for any given spot on the volume knob, the sound gets fizzy... maybe you've got something here I can try. It's great to have your engineering input around here.

...In single-ended amps I have become lazy. The last one I built (Vibrochamp) still hasn't had a single voltage measured in it while operating. I built it, checked the wiring and e-caps with a function gen/DVM, put in tubes and flipped the power switch. It's been going strong for about a week now and sounds groovin every time I turn it on. I used the stock bias resistor (as I always do on Fender cathode biased amps) and the amp sounds great right out of the box..

I am going to try and think you are smart and have reason to be confident in whats up - not lazy. Does my heart good to know that an engineer allows his ears and basic logic to be at work... the Human ear and mind is an incredible tool that many common folks possess (there are a few exceptions). Of course I expect you to measure it at some point and I'll be among the first to want to know how close you were. Too many years in the science field and you just can't live without asking "what does the data say"?

rj