What is the right current for a 300B at 400V and -80V grid bias?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.

jogas

... I am in trouble due to my low experience with tubes.
I am building a 300B PP 30W amp using the attached PP1.jpg schematic.
For some reasons I biased it like the attached PP2.jpg using a single regulator for each pair of 300B and a small single resistor (6.8 ohm) to ground them. (the original bias for each 300B was given by an individual AC winding grounded with a 10 ohm resistor as test point for the current.

Well, the PP1 is designed for 30W having 80mA per tube at -80V grid with 400V plate voltage. and the project specs say to adjust the current per tube at 80 mA having the V grid around -82 ...-80V.

The second amp is designed for 20W having 45 mA per tube at -80V grid with 385V plate voltage and the designer recommend to stay around 45mA adjusting the grid bias around -80V.

My PSU gives 394V and all the necessary current but, biasing the grids at -80 I can only have around 45 mA that are far away from the 80 mA originally given by the first amp specs.

What is wrong?

What is the right plate current to be expected with my 394V and -80V?

Any suggestion/explanation is more than welcome.

Giuseppe

Attachments

• pp.jpg
77.5 KB · Views: 767

PRR

Paid Member
None of those numbers smell quite right.

There will be w-i-d-e variation between WE-300s from different factories, different decades, and different amounts of wear.

http://www.mif.pg.gda.pl/homepages/frank/sheets/136/3/300A.pdf on page 5 shows WE's official plots, for what they are worth. (This is for 300A, but 300B is a different base-lock on the same tube.)

In practice: you apply your B+ and then adjust the grid voltage to whatever value is needed to set the desired current. In SET use, you often want to run the current up until the plate is not-quite overheating. In push-pull, you usually set for 50%-90% of the current that would exceed plate dissipation rating.

jogas

PRR you mean that assuming a max dissipation of 40W, having around 400V plate I can adjust the bias in order to have from 50 to 90 mA per tube?

What about the class A or A1 or whatever it will become changing the value of the current?
What is a generally accepted/used value and what is a reasonable ouput power without stretching the tubes too much?

Giuseppe

Gluca

300B curves say that you should have 100mA with 400V at the plate and -80V at the grid. May I ask how you measure the plate current?

A1: no grid current and plate current flowing even with no input signal

If you lower the anode current you move towards class AB, while decreasing the bias (i.e. going up to -0V grid bias) you enter class A2 with grid currents. I believe 300B does not like A2.

You can get more power out of the tubes in AB. Power depends on the output tranny BTW, I believe you can get +20W out of the 300Bs with this bias.

Ciao
Gianluca

jogas

Ciao GLuca,
I measured the plate current via the common cathode resistor and, to be sure I also measured the current with a mAmperometer in series to the primary windings of the two OPT.
Having 550 mV on the 6.2 Ohm cathode resistor (common to the two 300B cathodes) I calculated 88 mA....
The application notes of Svetlana from where I took the schematics say that the tubes must work at 80 mA with +400 and -80V grid....
Audio review from where I took the other schematic say that at 385V and -80 each tube should give around 45 that is what I get with 394V plate and -80V...
I understand that I can adjust any current value with the grid but... why this totally different current values?

Gluca

I have curves from WE, BTW it does not matter ...

If I looked at the first schema I'd see 80mA for each tube biased at 400V (minus the voltage drop in the tranny) and -80V at the grid and that point is on the WE plate curve. (I cant understand why they use such a large grid leak resistor, 221K, with grid bias?!)

Again if you set the bias at -90v with 400V at the plate you shoud have something like 45mA running through the tube. I don't know what that mag was looking at but its suggested bias seems to exceed the typical +/-10% difference from the 300B WE curves (385V,-80V, 65mA).

I am sure you double checked the bias voltage ... but, as I am reading a -230V on the rail feeding the bias (2nd schema) , probably a complete schematics with some values may help.

Ciao
Gianluca

jogas

Ciao Gianluca,
attached you will find the amp I built.
The bias I have chosen comes out from the second schematic, yje one wher they recommend 385/-80/45 mA that is what I get out of my amp wher I should have 80 mA per tube with -80 V grid. This is what I do not understand, since I am biasing 394V, -80V and I only have 40-45 mA per tube ...

Ciao
Giuseppe

Attachments

• no57fig1.jpg
58.6 KB · Views: 675

Gluca

That bias and set-up, from the last pic I mean, sounds familiar to me. May I ask if you measured the voltage diff between the grid and the cathode directly at the pins of the tube (or cathode resistor)?

Just to be 100% sure I would measure the resistance of each winding of the primaries' OPT, power up the amp and look at the voltage drop across them (ie. between B+ and the plate of each tube) ... this way you can easily measure the current for each 300B (separately).

Are you using svetlanas 300B or what?

Ciao
Gianluca

jogas

Hello Gianluca,

yes, I doubled and triple checked the values, they are correct.
To have 55-60 mA per tube I have to go down to -70...-72V.
I am using a quartett of Electro Harmonix 300B non gold grid.

Ciao Giuseppe

Gluca

I had a very similar experience

Ciao Giuseppe.

These values sound odd.

I had 4(!!!) ECC99 a couple of months ago running along strange and unusual curves... they were not giving the expected plate voltage/current for reasons still unclear. I believe it took a week to decide to throw them away. Now they are lying somewhere in my room. I understand btw that 4 ECC99 just cost as 1 300B by EH. Now I am using 6H30, in the same amp and with the same parts, with great satisfaction. I asked the shop for a refund but nope. Not even an answer. GET LOST!

I can suggest to test a single tube excluding all the remaining circuit and parts and see what happens for each of the 4 tubes. If they do not stick to the plate curves (allowing, say, a 20% tollerance), well if I were you I'd give your supplier a call and see if he can test them.

If the test tells you that the 300Bs are fine there must be something weird in your wiring.

Ciao
g

amperex

Denends upon the tube & operating life

Mesh plate 300Bs are best operated under 32 watts input with about 26 watts for longest life & good sonics. Solid metal plate tubes can take more input power, but I would limit it to 32-watts maximum. Best is to contact the tube manufacture.

CLS

Quite Basic, But Still Confused about 300B's Operation Points

Hi there,

I've searched a little & browsed through this thread just now, very informative but it did not resolve all my confusions. So I continue here.

Checked several 300B based SE amp designs, I found they're torturing the tubes! (including myself )

Let's review the specs, well, again:
Max anode voltage: 400V
Max anode current: 100mA
Max PD: 36watt

So, "ideally", no greater volatge & current should appear on the normally working 300B. At least that's the text book tells us, "no crossing the lines". But, let's check some examples:

Audio Note Kit-1: 425V on OPT, ~415V on anode, 65V on cathode with 750R resistor & flows 87mA
The Va=415-65= 350V
(However, as I marked this point Va=350V/ Vg= -65V on the plate curves, I saw about 120mA !! )

JC Verdier: 400V on OPT, assuming 10volts less on anode, same 750R cathode resistor, about 60V & 80mA
Va=390-60=330V
Again, on curves, Va=330V/Vg=-60V, more than 120mA !!!

There're several other design examples show higher than 400V on their anodes, 750R~1K cathode resistors for self-bias. I think they should be around the similar conditions at 330~350V/ 60~80mA range. (If I made any mistakes on the numbers, please kindly correct me.)

I can not pull the numbers together. What am I missing here? Is it simply "reality" vs "ideal" stuff mentioned earlier?

And there's more: let's draw the load lines through those points onto the curves, we'll see: when in full swing, more than 1/3 of the voltage & current would obviously exceed the MAX ratings.

While books say, never exceed the max ratings in ANY conditions.

Are they (or we) only considering "the center" of the signal waves? If only the "center" is within max ratings then it's OK?

Well, if so, that would be really harsh to the tubes, I guess.

I bought a 300B based SE amp a while ago, and modified it somewhat. With the oringal CR tubes, its HT is about 390V, modified 667R cathode resistor on the negative side of the filament & shows 56V on it (about 84mA). (Again, on the plate curves, I'm lost. )

Couple of days ago, I swapped the orginal CR tubes to EH ones, and felt it ran hotter. At first I just gave it a guess that EH tubes' current was larger in the same bias setup. But after some long enough operations (4 hours of more), it got too hot & gave an obvious worse hum on one channel. (I've not measured the actual current though. Before the hum worsened, it sounded pretty good to my ears.)

The hot-running condition & hum drew my attention back to the basic setup of the output stage, and made me review the data on hand, thus these horrible findings.

On the "drawing board", I see the curves begin straightening up when current is bigger than 40mA. If we took a relatively conservative 20mA as a minimum operation current & draw a popular 2.5k load line above that, we would get an operation point way higher than WE's recommendation (it'd be 350V, -75V, 100mA vs WE's 60mA). What's happening? It's already at the max current rating on zero signal & almost touches the max PD!!

And, IMHO, 4~5k load on 300B SE should be a better choice when low distortion & high linearity are at the first consideration. It's just less output power. So, it's all about trade-off, huh?

What's your favorite operation point on your 300Bs? How do you like "torturing" them?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.