trinityamps.com

hi
i have an interogation about my cathode resistor rating
on the sII layout the value is 125 ohm on the kit a 120 ohm was provided

anyway: i went to this page http://www.webervst.com/tubes/calcbias.htm
on the Calculate Plate Dissipation In Cathode Biased ... calculator i 've input my voltages

V4/V5
P3 10.7 cathode
P7 328 plate
P9 318 grid

and the result was very strange for me
according to the calculator my plate current is 42ma and my plate power dissipation is 13 watt this is very hight value
i think the amp is biased very hot

according to this calculator i should replace my 120 ohm cathode resistor by an 180 ohm these should give me 26 ma of plate current and 9 watt of power plate dissipation

these value seems better for me

have you any idea or advise
thanks
jerome

nobodys have an idea about a 180 ohm cathode resistor instead of the 120 ohm stock ?

Original value is 125. I've seen 100 to 140 in there. You can increase it if you want, but I would not go more than 140.

The cathode resistor along with the Plate resistor control the gain of the tube stage. Typical values are anywhere from 100 ohms to 10K. Smaller values = more gain. Personally, I like 120.

Measure the voltage dropped accross the resistor, square it, then divide by resistor value in ohms. Divide by two for two EL84s. See what you get.

_________________
Stephen
Web: www.trinityamps.com. Facebook: facebook.com/trinityamps. Twitter: @trinityamps

You should be OK with the resistor as is.

Also, you can't really compare cathode amp biasing with "fixed" biasing rules of thumb. It is not unusual for a cathde biased amp to be biased close to 100% of expected power plate dissapation.

Due to how a cathode biased amp operates, it will look as if it is biased very "hot" compared to a "fixed" bias amp. If I remember correctly, a cathode biased amp actually momentarily bias' colder when it gets hit with a large signal, the bias changes and reacts to the input signal where in a "fixed" bias amp, it does not change and you need to bias colder to prevent damage.

I hope that makes sense and that someone else can explain it better.

this formula should give me the tube power dissipation ?

Yes, at the power you are running it at that time.

_________________
Stephen
Web: www.trinityamps.com. Facebook: facebook.com/trinityamps. Twitter: @trinityamps

That is a somewhat approximate calculator. In my opinion it's best to calculate the dissipation yourself.

- Measure the voltage across the cathode resistor and divide by the value of the resistor. Then divide that by two to get the individual cathode current.
- Measure the voltage across the screen grid resistors and divide by their value, to get the screen grid current. If it's only a single shared screen resistor, then also divide the value by two.
- Subtract screen grid current from cathode current to get anode current.
- Multiply the anode current by the difference between the anode and cathode voltages (ie true anode voltage) to get the actual power dissipation.

These days electricity supply voltages are higher than when 18W amplifiers were first designed, so the DC voltages are also somewhat higher. Also valves were cheap and nobody cared if they had to replace their Mullard or Telefunken EL84s every couple of years.

Personally, I recommend 150 ohms as a minimum value for the EL84 cathode resistor in 18W amps, following the advice of DIY amp guru (and semiconductor physicist) Paul Ruby. You will therefore also see this value used in the 18W Lite IIb and SuperLite TMB designs. One of my amps is using a 180 ohm resistor at the moment, and in amps with higher voltages you may find as high as 220 ohms being used. IIRC the Watkins Dominator (which the Marshall 18W design came from) also uses 220 ohms.


Yes, that's about right. In a cathode-biased amp, as cathode current increases, the bias voltage also does, so biasing the power tubes colder. There is just one problem with that - if the bias voltage rises too much, it can force the tube(s) into starved mode, so the distortion sounds rough and also some crossover distortion can occur. So to get around this problem, you either have to bias fairly hot, even close to 100%, or else you bias colder (say 75-80%) and use a very larger cathode cap in the 1000uF to 2200uF region to reduce bias voltage fluctuations. The huge cap can also help prevent "boomy" bass response. Some people prefer to use more inexpensive, modern-production valves and bias them hot. Others may prefer to use NOS valves and bias them cooler. The choice is yours.

_________________
Great tones come in small glass jars!

Thanks for the explanation zaphod, now I know why my amp seemed to starve during heavy rocking.

Just as you strat777 I was worried about too hot bias so I changed my cathod resistor. The new resistor might have been the cause of my issue.

I did a lot of reading on the 18watt forum when I built my SIIIv6 so I tried zaphods huge cathod trick (1000uf) and it helped on the starving thing . But it made my amp a bit dull sounding . So I then tried the small and huge cathod trick, 1000uf parallell with 1uf film cap. And the high end was back. But I wanted more mid definition as well and increasing to a 2,2uf film cap was perfect for that. It gave the amp more pick attack.
So if you try huge cap experiment with a small as well.

Now I'm thinking of using one set of tubes with huge cathod and the other with the original hot bias. But I can't decide on bluesy el84 or 6v6...

ok thanks for this very helpfull information but i have one little probleme with this




could you tell me where i could measure the screen grid voltage
i've tried with my pin 9 ( 318V divided by 100R divided by 2 )
but this seems to be not the right place to take the voltage it give me 1.59 amp

thanks for help

_________________
Stephen
Web: www.trinityamps.com. Facebook: facebook.com/trinityamps. Twitter: @trinityamps

Measure the voltage between the ends of the 100 ohm resistor. That will tell you what the current through it is.

_________________
Great tones come in small glass jars!