How to calculate load resistance from tube data?

[ALBATROS1234]

I like to experiment with unique designs and its sometimes fun to just hook up stuff and see what happens. My question  is when i want to figure out what the primary impedance of an output transformer should be close to for a power tube that load resistance is not given on the datasheet. As an example on the 6v6 the rca manual gives you a range of 5.5 to 8.5k primary impedance depending on plate voltage. The plate resistance ranges from 50k to 80k. is it as simple as dividing plate resistance by 10 to get in the ballpark or do you have to consider other variables such as transconductance? For instance i have some 6hb6/6ha6 power pentodes i want to experiment with . There is no load resistance listed. Plate resistance is 24 to 28kohms so do i divide by 10 and get 2.4 to 2.8k primary or is there something else to the formula? I have purchased several tube lots in order to get some tubes i wanted. But i ended up with literally hundreds of tubes that i dont have much use for. But digging through you find tubes like the 4 nos rca 6hb6 power pentodes which look an awful lot like a like an el84 with plate dissipation of 10w and 350v max plate. I dont get why every power tube doesnt have included in datasheetss load resistance.i would very much like to see if these bad boys sound as good as they look. there are dozens more types i have as well , like some 6cu5s etc. Inquiring minds want to know. Thanks gents. P.s. good luck to everyone in these trying times

[shooter]

is it as simple

nope
there are 2 Different values
plate resistance and Load resistance.
in PA's we typically only care about the load.  That varies with plate volts, tube type, speaker load............

for load, I haven't found a data sheet yet that didn't offer up some "example"
you can use ohm to ballpark, if you have Ip and Vp

[kagliostro]

Talking of SE

Take the value of B+ you want to use (consider the max Va on the tube datasheet to establish it), on the datasheet give a look also to the max plate dissipation and compute a ...... say 90-95% of it

Output Load = (B+ x B+) / Established limit for dissipation

Z = Va^2 / Pa

Assuming you have 250V B+ and a dissipation max 11W (the max you want on the tube based on max allowable for the specific tube)

Z = (250V x 250V) / 11W

Z = 62.500 / 11  Z = 62,500 / 11

5.681 Ohm Load 5,681 Ohm Load

Franco

[shooter]

 
K, you slip decimals like I do 

[PRR]

> I dont get why every power tube doesnt have included in datasheetss load resistance.

_Audio_ load? Why should they? There are a dozen very good Audio power tubes fully specified and totally functional.

6HB6 is a _TV_ tube. Video Amplifier. It has large gain because video is hard to amplify. It has a huge cathode (and heater) to get that large gain.

You can buy 6V6 or EL84 for, what, $20? The 6HB6 is $1, bargain!! Or is it? It will take some thought to suss-out a "good" operating point. From a cold start, I'd estimate $50 of brain-time. Actually I've looked at this before and could quote you $25 for a re-sussing and the write-up.

Here's a free version. *Assume*(?) the "Average Characteristics" are also a happy operating point. 250V, 40mA. As said, a good load is probably near V/I or 250/4 or 6.25k ohms. The sheet bias point is 100 ohms cathode resistor, 40+6.4mA= 46.2mA, so 4.6V grid bias. You will need up to 4.6V of audio grid swing (MUCH less than most audio tubes).
http://www.nj7p.org/Tubes/PDFs/Frank/127/6HB6.pdf

Some checks. Yes RL should be much less than rp. But "10" is just coincidence. Across many types you could come up with 2 to 50. Anyway a Pentode has TWO plate resistances. On 6HB6, above Vp=40V it is about 50k. But under 20V it is near 200 Ohms. If you mess with Ohms Law both ways, it seems that the load should be the split-difference of the two plate resistances. But this optimum is very broad.

You should also go to the curves and see if you can get double the idle current on peaks. In fact at Vg2=250V you can get triple the idle current! Or you can turn Vg2 down to maybe 180V, and need less bias and drive.

Read and understand the above explanation. How long does it take you? How much is your time worth? You may be cheap but you are not worthless. If you are going to build a million amplifiers, $50 of brain-pain to use a $1 tube makes sense; not if you only build one and can read the datasheet of a $20 tube in $2 of brain-time.

[DummyLoad]

K, you slip decimals like I do 

in euro for math: decimal is , and comma is .  but you knew that and yer just jerking K's chain, right?


--pete 

[kagliostro]

Z = 62.500 / 11   Z = 62,500 / 11

5.681 Ohm Load 5,681 Ohm Load

Better now ? 

Franco

[DummyLoad]

Z = 62.500 / 11   Z = 62,500 / 11

5.681 Ohm Load 5,681 Ohm Load

Better now ? 

Franco

at least we can agree the brits drive in the wrong side of the road...   


--pete

[shooter]

 
I was just, jestin 
and the Brits would disagree

[BrassElephant]

Talking of SE

Take the value of B+ you want to use (consider the max Va on the tube datasheet to establish it), on the datasheet give a look also to the max plate dissipation and compute a ...... say 90-95% of it

Output Load = (B+ x B+) / Established limit for dissipation

Z = Va^2 / Pa

Assuming you have 250V B+ and a dissipation max 11W (the max you want on the tube based on max allowable for the specific tube)

Z = (250V x 250V) / 11W

Z = 62.500 / 11  Z = 62,500 / 11

5.681 Ohm Load 5,681 Ohm Load

Franco

So this is awesome, I was gonna start a new thread asking the exact same thing. This explanation is dumbed-down enough even for me!

How do we take this SE math and apply it to PP operation?

[jjasilli]

Here's  a load calculator for a pentode:  https://www.vtadiy.com/loadline-calculators/power-stage-calculator/


Hopefully it's helpful

[ALBATROS1234]

thank gents for the help, and p.r.r. i understand what you are saying, and i am not trying to go into the amp manufacturing buis, i am a guitar player/ tone geek who plays with tubes. i have noticed that tubes have variable tones even within same types, sometimes within the same brand. my experiments have yielded that there are great and interesting tones to be produced from tubes never intended to be audio tubes. this is a hobby for me, i am not concerned about how much time it takes me to succeed or fail to produce desired results from this tube or that tube. i guess since i have only been screwing with tubes a few years its still exciting and interesting to experiment and play with values and different tubes and different methods like the interstage tranny p.i. or whatever. i just love that raw vintage tube tone i get from a simple circuit and a few tubes with some voltage. i expect to fail most of the time but each failure refines my understanding. there are surely holes in my knowledge although i am not a complete greenhorn being an electrician plus having 2 years of electronics school and around a decade of electronic repair experience. i like electronics, i love music, i enjoy tinkering. plus i am shut in at home and i got a pile of tubes with potential and all the time in the world. i appreciate any and all advice and help thank you sir.

[ALBATROS1234]

kagliostro does plate resistance play no part in this? i imagined there would be a formula which include operating voltage,plate resistance and transconductance at least to get to reflected load value.

[shooter]

reflected load value.

the datasheet gets you close enough to solder.
effective load resistance in this example;

[kagliostro]

kagliostro does plate resistance play no part in this?

Unfortunately I'm not so skilled, I borrowed my math at Merlin's Web Pages

SE
http://www.valvewizard.co.uk/se.html

PP
http://www.valvewizard.co.uk/pp.html

Franco

[ALBATROS1234]

shooter i am not talking about common tubes, i can look up on the datasheet and read load resistance of 12000 ohm which is 12k ohm so my primary is 12k. i am talking about perfectly good tubes that would and could make a great sounding guitar amp like i mentioned above the 6hb6.i got 4 nos rca of this type. i could not help but notice they look identical to other power pentodes aka the 6bq5/el84 except there is no listed load resistance on any data sheet because it was intended to be a vertical deflection amp in tv sets. another tube that was intended to be a vertical deflection amp is a 6cz5 which is damn close to a 6973 (same pin out and sounds practically identical) except for slightly higher transconductance. the 6v6 was used in tv sets in that use as well. my point is theres a ton of perfectly good tubes in true nos condition that most people like us ignore. i simply wanted to see how i could calculate load resistance from the given data on tubes where no "load resistance" is given. i assumed plate resistance would be a factor, as well as operating voltage and possibly transconductance. there must be a formula. i just need to find it. i am presently doing it with a sledgehammer by just wiring socket adapters and seeing how they sound in some of my amps i have built, if they sound shitty i then check voltages and tweak bias if necessary or try a different amp with different primary on the o.t. sometimes i have to pad the grids with larger stoppers because grid impedance is different. idk it just seems wasteful to have all these perfect condition usa made tubes collect dust. just my opinion, i hoped someone could help. p.r.r. did help some. i still think theres a formula somewhere that i can punch in values and get my load resistance on tubes where it is not given.

[kagliostro]

I've find this, don't know if they can be of some interest









Franco

[shooter]

My current build is with 6K6, horizontal TV tube, I now know why it's not highly used in amps, I also know it sounds pretty good.

there must be a formula.

there is, sorta;
If the data sheet shows suggested plate volts and tube current, you have load
R(load) = ~~ Vp/ tube current.

If the datasheets say NOTHING about audio, somebody(s) figured out it probably isn't worth the paper

[Ed_Chambley]

Here is a link to one playing these tubes.  Probably SE Stereo, but may provide a trail.

[shooter]

 
I spent a lot of surf,

Here is a link

,time on orphan tube examples 

[Ed_Chambley]

RCA DATA SHEET states A1 amplifier, 250 Plate/250 Screen 24000 ohms which would make try somewhere near 5k reflected, but remember I am a hack.


The video I posted, if you go to youtube to watch there is link under it to buy parts.  I just cannot real any Asian text.

[kagliostro]

Many times I go this way:

I do a search for images with google

usually I search for xxx(the tube of interest) amp schem

Then I perform a visual search among results hoping to find a link with info I can be interested on

this way I've just find this

https://antiqueradios.com/forums/viewtopic.php?f=12&t=273358&start=2000



Some time you find a good project with good data (read the page link with attention to guess it) other time no

Franco

[ALBATROS1234]

ok thats some good clues thanks guys.so btw 5k to 8k primary, i tryed it on 5k and the result was fair. that last schematic gives me a good clue especially on biasing resistor. i tryed 500ohm,(already in the amp) 250ohm and 150ohm. this says 82ohm so perhaps i will try a 100ohm. been mostly working on restoration of a 1935 rca 225 console radio which i will be putting lacquer on in an hour or so. i will get back to my 6hb6 later on to see whats what.

[tubeswell]

How do we take this SE math and apply it to PP operation?

In simple terms, the plate-to-plate load (across the OT primary) will be the same as a SE load from plate to B+ for Class A operation.


e.g. Z for a centre-biased 6V6 at 310V on the plate in SE is Z=310/(12/310) = 8k in Class A


So 2 x 6V6 at 310V in parallel Class A want 8k plate-to-plate (which is 4k for 'each half' of the primary. There are 2 halves* in the primary, so these 'add up' to 8k plate-to-plate)


* each half of the OT primary sees current, albeit that under signal conditions, one half sees 'negative current' going in the opposite (turns) direction to the other half, so the overall effect is the same as having 2 x positive current going in the same turns direction across the whole primary. Hence, the 2 halves of the primary are 'side-by-side' in terms of overall Pri-Sec (AC) induction (although in DC terms, the whole winding is one piece of joined up wire in series from one plate to the other plate). In AC-operation (under signal conditions) the current enters/exits the primary winding at the centre tap, and in one half, travels as negative current going in the opposite turns-direction to that of the current in the other half of the primary (even though the whole primary is wound in the same turns direction)


Class B is different, because when the output tubes start to run in Class B, one half of the primary is 'off', and so the Primary:Sec turns ratio is halved (and therefore the Pri:Sec load resistance is 1/4, because load resistance is the square of the turns ratio).


So much for the theory. In reality, you need to be careful about screen voltage, because if the load is 'too high' (i.e. too-shallow-a-load-line) and the screen voltage is 'too-high', then the screens will be forced to dissipate more than is desirable during heavy signal conditions.  So it can be better to run the load a little lower (i.e. steeper line) than the calculated Class A optimum load, or bias the tubes a little cooler, or lower the screen voltage to get the knee of the Vg=0 grid curve at or below the load line.