• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Figuring plate resistor values or tubes 101

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This may seem dumb to some of you that have been working with tubes for years. What is the correct way of figuring the value of a plate resistor? Is there a ratio of plate voltage to power supply voltage? I have seen the term plate resistance under the "Characteristics and Typical Operation" area of tube applications.

Example: A 5687 is listed as having a plate resistance of 1.5k at 120volts and a plate current of 36 Milliamperes.

What would be the minimum plate resistor value? What would be the minimum B+ voltage before the plate resistor? What would the target cathode voltage for said plate voltage? What would be the target plate voltage?

If this tube were operating at 140-150 volts on the plate with a cathode voltage of 5.2 volts or about 11mA would it sound different at say half the plate voltage and half the cathode voltage?

Any help here would be appreciated.

Joe
 

G

Member
Joined 2002
Hi Joe. While I'm certainly not an expert the rule of thumb that I have followed is to use a plate resistor of at least three time the plate resistance of a triode. The plate resistor is what makes the tube work. Look at it this way. Picture a pair of resistors in parallel. If one resistor is 10 ohms and the other is 1 ohm then 90% of the current(signal) is going to flow through the 1 ohm resistor and only 10% through the 10 ohm resistor. The plate resistor determines how much signal is going to the next stage or component in the same way. If you don't use a big enough plate resistor then all of your signal is going to go through the tube straight to ground and no signal will flow to the next stage. The value of the plate resistor determines how much signal will flow through the tube to ground and how much will flow through the next stage. Remember that that is the AC function of a plate resistor. The DC function is to drop the B+ to the appropriate value for the tube at a given level of current. Let's take the 5687 circuit for an example. The circuit that I am using requires 150 volts on the plate at 10 mA. Looking at the curves on the datasheet I see that in order to have 10 mA current flow with a plate voltage of 150 volts I need a negative voltage (or a voltage drop across the cathode resistor) of about 6.8 volts or so. This means that I will need a cathode resistor of 6.8/.01 = 680 ohms. So now you have the operating points of your circuit but you still need to determine the plate resistor value. I like to drop half of the B+ across the plate resistor and the orher half across the tube/cathode resistor. Which meand that my B+ is going to be 150 +150 = 300 volts. Now to drop 150 volts at 10 mA I need a resistor value of 150/.01=15000. So my plate resistor needs to be 15K. These operating points will give me a pretty good circuit. So these are the operating points I am using for a 5687. I'm not in Frank's or Brett's League but I hope that I helped you instead of confusing you. I'm sure that they will clarify what I'm trying to say.
 
It really helps to have an understanding of plate curves. Then all becomes clear and easy to calculate.

In the mean time, a plate supply resistor should be more than 3 times the plate resistance at your chosen operating point (plate voltage, current, grid voltage), and should drop between half and 2/3 of the supply voltage for best linearity, leaving the other 1/2 to 1/3 of the supply voltage for the tube.

Using published operating points isn't usually a good starting point because they might've been drawn up for say, transformer coupling. The current or voltage may be too high to be useful for an RC-coupled stage.

Oh yeah, it sounds like you need to understand the basic parameters of tubes. Rp = Ra = Plate Resistance, the effect a change in voltage has on current. Say you increase the plate voltage 20V while keeping the grid voltage the same; if the plate current increased by 5mA, the plate resistance at this average operating point is 20/.005 = 4kohms.
Transconductance = Gm = mho (unit of conductance) or S (Seimens), usually in micro- or mili- units (3200µmhos = 3.2mS). Say you increase the grid voltage by 1V, and plate current rose 5mA as a result (plate voltage is kept constant). Gm = .005/1 = 5000µmhos.
Amplification factor (symbol "µ" (Greek mu)) - Say you took the same change in grid voltage, +1V, and adjusted plate voltage so the current is the same as before [the grid voltage change]. Say the plate had to be lowered by 40V to get the same current. µ = 40/1 = 40. Usually not specified for tetrodes and pentodes because they have the unusual characteristic of having a plate resistance much higher than the load resistance; as a result, mu is very high (often over 1000) and actual gain is very load-dependent, making it a relatively useless spec.

HTH.

Tim
 

G

Member
Joined 2002
What he said.:nod: I'll try an analogy that may make it easier to understand. The same one my teacher used when describing FETs. Look at a tube as a water valve. The more you open the valve the more water flows through it. How much flow through it for each turn of the handle depends on the pressure behind the water. If you open the the valve all the way then the flow through it increases but the difference in pressure across the valve decreases. Negative voltage on the grid is like the hand that turns the valve. While the supply voltage is like the pressure behind the water and the current is the rate that the water flows. The negative grid voltage applied to the grid is like the hand that opens or closes the valve. The more the valve is opened the higher the flow of water through it and the less the difference in pressure across the valve. The more negative voltage applied to the grid the more current it allows through the tube and less voltage is dropped by the tube. This concept is called Ohm's Law and the eqaution for it is E(voltage)=I(current)xR(resistance). This equation is the basis of electronics. Voltage and current are always inverse to one another if the resistance is static. More voltage means less current and vicey versy;). By now you should have a screaming headache and be reaching for the good ole Ibuprofin. Hope this also helps.

G
 
Thank you very much for the answers on the plate resistor selection and power supply size. It makes a heck of a lot more sense now.

Back to tube regulators now... HELP!!

I was trying to use an 0D3 in the power supply of my 5687 line amp only to find out that it won't give me the plate voltage that I require after using a suitable plate resistor.

After looking at Bas schematic he used a 0D3 and another regulator and ended up with a supply voltage of around 250volts. I would like to end up with a supply voltage before the plate resistor of around 240-250volts. I would then use a plate resistor of 10K and end up with a voltage on the plate of around 130volts. I believe this would be very close to the ideals of about half the B+ voltage.

This would make a current draw of about 11mA per section plus the 5mA per regulator tube to turn on? Well, I can't get it to turn on and run the two regulator tubes. This just has to be black magic because it just won't work for me.

I would really like to use a plate voltage of around 130-140 volts and a cathode voltage of around 5volts because I heard a line amp with this and its unbelieveable. I question that I would be able to get the same sonics and feeling out of a lesser B+ supply.

Part 2

I purchased 2 Hammond chokes, they were 156c's. Would they be suited for usage as a plate chokes? If so how would you use them with the setup I want to make?

Any help here would be appreciated.

Joe
 

G

Member
Joined 2002
It sound to me that you are trying to reinvent the wheel. I think that the original circuit that you are trying to build is here:

http://diyparadise.com/preamptutorial.html

You'll notice that there are no diodes in the power supply nor does there need to be for the circuit to sound good. I've never heard a tube circuit that used a regulated power supply but I have heard half the people say they sound better and half the people say they sound worse. Maybe you should build the original and then modify it and decide for yourself.
 
You are correct I am building that circuit. I do however want a regulated supply. I am convinced that they do sound better because I have heard both a regulated version and a standard version before I started putting mine together. I am sold on regulated supplies. Note: I have also heard a version without chokes in the power supply and I have heard a version that uses a battery to supply cathode voltage.

One thing to consider here... If the solid state amps (the better ones) use regulated supplies, current mirrors, stepped supplies, and other assorted goodies and they sound better the same should hold true for tubes. I for one believe it does.

Before I am done with this project there is no doubt that it will have dual regulated power supplies. This I learned from years of playing with solid state.

Maybe it will have plate chokes also or CCS.

I am also sure of another fact, it will cost more money and that extra money spent will be reflected in the quality of the sound.

Joe
 

G

Member
Joined 2002
You are correct Joe. It's your money and your time. If you have heard both versions and you like the regulated one then that is the one you should build. You are fortunate in that you know what you want and what to expect when you finish. I tend to like more basic designs and I definately am content in my blissful ignorance. I meant no offense. Enjoy.
 
diyAudio Senior Member
Joined 2002
Hi,

My personal feeling is that the design is a good one and I think it can be made better by spending a $100 or so on mono supplies and regulation.

Agreed...but why not go all the way and use a real regulated suppy?
A VR tube can only do so much.

If you have 50V spare voltage on the PSU a dedicated, one per channel regulator can be added.

You'll need a spare 6.3V heater supply xformer, a seven pin + a noval socket, an ECL85/805 and a 85A2/5651A per channel, some resistors and caps and you're all set.

An example:
 

Attachments

  • psu_245vdc.gif
    psu_245vdc.gif
    9.9 KB · Views: 376
HEY-Hey!!!,
I have built a VR tube shunt regulated linestage, 5687 amp tube and Cascoded DN2540's for plate loads and current regulation before the VR string.
I have done direct comparison to the identical amp with a series pass tube sort of supply like you have shown. Single bypass of the VR sting with a TFE .056 uF.
The shunt reg sounded better to all those who listened to it so far.
Same plate loads(DN2540 ccs), same bias method and point, same output coupling caps. Final construction done specifially for the purpose of eval the PS.
I have actually considered the VR string for use with a choke input supply. The string will strike and conduct( even if barely) as long as the amplifier tubes don't come on and conduct more than critical current( and then some, actually). THe VR string wil conduct just enough to keep things from getting out of hand and go out when the amp tubes conduct.
regards,
Douglas
 
diyAudio Senior Member
Joined 2002
Hi,

I have done direct comparison to the identical amp with a series pass tube sort of supply like you have shown. Single bypass of the VR sting with a TFE .056 uF.

Generally speaking, a shunt reg offers better sonic performance ove a series type reg...at the expense of effeciency.

If you look at the circuit posted above you'll notice the big 330µF cap behind the reg.
You'll also notice that the reg isn't really there to perform as a reg but to provide high isolation from the main PSU and by the same token, from the mains supply.

Add a choke as anode load or a CCS and tell me what you hear.

If you hear incredible dynamic range, pinpoint holographic imaging with this kind of topology, you'll know you're onto something special and you'll have the op points for the valve spot on.

You can push it even further but the return just isn't proportionate to the outlay and hey...I'd like to keep on posting for a while longer, right?

Assuming source material and speakers are up to it of course.

If CD players would adopt the same thinking in PSU design we might, just might start to enjoy minced music a little more.

Cheers,;)
 
Hey-hey!!!,
I never would have thought I'd post something like this, much less admit it, but on the other side of two cascoded CCS, I could still hear the rectifier tube. I went through a few and settle on the Sophia copy of the WE274B, in perf anode, globe version. Gary pimm measured those CCS at tens of G Ohm and there are two in series for a given channel. Doesn't make reasonable sense, does it. I also tried some Auri caps for main PS, then went to motor run oilers for improvement.
OT, isn't the Balloon Festival in St.Niklaas somming up soon?
regards,
Douglas
 
diyAudio Senior Member
Joined 2002
Hi,

I never would have thought I'd post something like this, much less admit it, but on the other side of two cascoded CCS, I could still hear the rectifier tube. I went through a few and settle on the Sophia copy of the WE274B, in perf anode, globe version.

I hear people say things like yourself...with te stepped PSU, for that's what it is, I doubt you would.

I know I don't don't...not that that is anything to go by.;)

I also tried some Auri caps for main PS, then went to motor run oilers for improvement.

IME, it's a waste of money...the oilers and Auricaps are just too good in that application....after the reg, yes only the best is good enough for Oscar Wilde.

OT, isn't the Balloon Festival in St.Niklaas somming up soon?

Woa...you found my resume on the net or something?
Yep, it's early September usually...

Geez...I hope my firewall is still intact...:scratch:

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