ELR (effective load resistance) Question

[Toxophilite]

From checking out tube datasheets
For an amplifier running in AB1

it seems like the ELR of 6l6 tubes is 3800 ohms  with a plate voltage of 360vdc
While  6L6Gc are 5600 with a plate voltage 450vdc


Does it change the effective ELR of the 6L6Gc if you run them at 365vdc on the plates like in a vibrolux 6G11?

[jazbo8]

I don't see why not... below is taken from the GE datasheet (please note the screen voltage also):

[Toxophilite]

Cool
I was looking at a few different data sheets and there seemed to be some with more information than others..One thing that confuses me about the GE chart shown is the the 1st column which seems to share many of the values (other than max sig screen current and max signal plate current)
yet there is a huge difference in the ELR and maximum signal power output

[jazbo8]

One thing that confuses me about the GE chart shown is the the 1st column which seems to share many of the values (other than max sig screen current and max signal plate current) yet there is a huge difference in the ELR and maximum signal power output

To see why that is, you need to draw the load lines and compare them, but those figures would suggest that 6.6k is a better load match than 3.8k using the same operating voltages.

[HotBluePlates]

it seems like the ELR of 6l6 tubes is 3800 ohms  with a plate voltage of 360vdc
While  6L6Gc are 5600 with a plate voltage 450vdc

Does it change the effective ELR of the 6L6Gc if you run them at 365vdc on the plates like in a vibrolux 6G11?

The output tube doesn't have "ELR" as you've described it. In fact, if you look at the plate curves for the 6L6, above some low plate voltage, and individual gridline turns and runs mostly horizontal, indicating a very high internal plate resistance. But we really don't care what a pentode or beam power tube's internal resistance is; it doesn't matter the way a triode's internal plate resistance does.

Instead, think of a pentode or beam power tube as a "voltage-controlled current source". Applied grid voltage causes these tubes to pass more/less plate current, and the differing position of the grid voltage lines on the sheet show this. Also recall this when noting that you generally don't see Mu or plate resistance for the tube listed, but you always see Gm for these tubes. For a 6L6, 6,000 micromhos is typical, but Europeans specify this as 6mA/V (same quantity with a different way of expressing the units).

You could have a short hunk of wire with an ammeter between the power supply and 6L6 plate, effectively 0Ω load resistance. And for a changing grid voltage, you will see a changing plate current. But that's not useful in an amp. You need a load impedance through which that plate current flows so that, in accordance with Ohm's Law, a voltage can be developed across that load impedance. And Power = Voltage * Current, so now you have power developed across the primary winding of the OT, to be passed to the secondary.

What load impedance to use? Depends on the supply voltage, the limit of plate current for the tube used, and how much power you need.

If (especially screen) supply voltage is low, plate current swing will be low and you might need a bigger load impedance to develop a bigger voltage swing to hit your power target; this accounts for higher load impedance at lower plate voltages.

If you don't need as much power, you can keep supply voltage and idle conditions the same and use a smaller load impedance. Smaller load "R" when plugged into Ohm's Law (Volts = Current * Resistance) yields smaller plate voltage swing. Power is then smaller because (Smaller Volts) times (Same Current) equals Smaller Power. This matches the somewhat lower power output for the 3.8kΩ condition compared to the 6.6kΩ condition for the same supply voltage.

In the third condition listed, the designers went further into class AB. The supply voltage, idle current and bias went up. Idle current is ~32% more than the other conditions. But load impedance went down compared to the first condition, which allowed a max-signal plate current 59% greater than the first condition. Power output nearly doubles.

[HotBluePlates]

Point of all that is there is no "one calculation" or measurement* to find best load impedance for a set of output tubes with a certain supply voltage. There's a starting point for class A of taking max plate dissipation and dividing by your proposed supply voltage, but then that has to be jiggered a bit because it assumes the tube is an ideal device.

Instead, you make a first approximation, calculate plate voltage and plate current swings, plot loadlines, estimate distortion, see if any maximum gets exceeded, think about whether it could be better. You check if the tube would overheat at some point of the signal cycle. Then you pick a direction (higher or lower load impedance), rinse & repeat. And a third time. Maybe a 4th, 5th time. Eventually you settle for a best load for your tube, power supply (voltage and current) capabilities and power output needs.

The key takeaway is the tube could deliver some maximum amount of power if a suitable power supply allows it. At the same time, you have to balance the reflected OT primary impedance against supply voltage, desired plate voltage swing and tube plate current capability, and pick an impedance which gives the best balance of a lot of competing factors. You're not really matching the OT primary impedance to any resistance/impedance within the tube (for output tubes, anyway).



* Actually, there are power meters which allow the operator to present different loads to an amplifier, allowing the operator to see quickly which load yields the most power. I've got one, and it can measure power up to 20w, though with an appropriate pad circuit between the amp & meter, power up to 200w can be measured. But highest power may not equal lowest distortion, or the particular load could cause some limit of the tube to be exceeded, etc. Simple as the meter is, an operator needs to know quite a few things to avoid damaging the meter or the amp.

[Toxophilite]

THanks
Whilst I don't understand the bulk of what you are explaining(due to my limited knowledge). I do understand that there are many factors at play and that it is not a question with a simple yes or no answer

[jazbo8]

HPB is talking about the effective plate resistance not the effective load resistance that I think you were asking about, the two parameters are related but not the same.

[Toxophilite]

I did wonder about that, thanks

[PRR]

> Does it change the effective ELR

Optimum load, not "effective" load.

The total power output is (usually) soon limited by dissipation problems. So higher voltage *must* use lower current to stay safe. If voltage is up and current is down, the load impedance *must* be higher.

Original-recipe 6L6 claimed 425V but reliability issues soon changed this to 360V. 6L6_GC_ is really a very different tube (but works in all 6L6 sockets) with higher voltage *and* dissipation ratings.

I do not understand the question "plate voltage of 360vdc... if you run them at 365vdc". 360 and 365 are the SAME for any practical purpose. We are not going to custom-wind a 6.78K (3k8 is wrong) transformer just cause we sit 5V higher. The "optimum" load is not THAT critical, not by a long shot. Also the marked impedance assumes an ideal final load; our "8 Ohm" speaker will be 6 Ohms and 50 Ohms at different points in the audio range (but intended to suck best if you work the numbers as-if it were an 8 Ohm load).

Also your "365V" may be at no-signal? What is it at FULL-signal? I'd bet it is closer to 360V.

Also when I move from my old office to my new house, all voltages rise 15% (different kinds of wiring problems in each place). So are you figuring for your workshop, Pat's Posh Pub, Dave's Dive, or the concert in the park (very long extension cord)?

If you want to go from 360V to 400V, that's 10%, which is barely enuff to matter. If voltage is up 10% and the bottle is now too hot, you want to reduce the current 10%. load should go up by factor 1.21, to like 8K. However 400V strictly requires something other than original 6L6; anyway today all you get are 6L6GC-like jugs. The higher Pdiss means you "can" run higher current also. Actually much higher, which is why the GC condition shows a much higher voltage *and* a slightly lower impedance, for a much higher power output.
_____________________________________________

> the 1st column {of the AB1 conditions}which seems to share many of the values
> yet there is a huge difference in the [/i{suggested load resistance} and maximum signal power output

Now look at the AB2 conditions. You find the same two load resistances, 6k6 and 3k8, but now very different power outputs.

They are showing that a good-fit in AB1 does not mean a good fit in AB2.

This is now only Historical Interest.

In 1938 it was still common to use a driver transformer. This can drive G1 positive and get a lot more current through the tube, allowing higher power output into a lower load resistance.

A good driver transformer is about the toughest job around. Everybody switched to R-C coupled tube drivers around 1939-1949. Nearly no guitar amps run AB2.

So 360Vp 270Vg2 and 3k8 loading is a poor fit for any amp you will work on.

(Take G2 higher, and now you can approach the "AB2" conditions without actually getting into AB2. 5F6A is 400V P and G2, 4K load, 40 Watts out-- but 400V on G2 was illegal on original 6L6, only became OK with 5881 and the 6L6GC.) [/i]

[Toxophilite]

 
"I do not understand the question "plate voltage of 360vdc... if you run them at 365vdc". 360 and 365 are the SAME for any practical purpose. We are not going to custom-wind a 6.78K (3k8 is wrong) transformer just cause we sit 5V higher. The "optimum" load is not THAT critical, not by a long shot. Also the marked impedance assumes an ideal final load; our "8 Ohm" speaker will be 6 Ohms and 50 Ohms at different points in the audio range (but intended to suck best if you work the numbers as-if it were an 8 Ohm load). "

I think you misundertood that question and likely it is my fault,
my question was NOT AT ALL about the difference between 360 and 365 volts (I know the difference is very small and the plus and minus 10% or sometimes 20% tolerances that tube gear and associated components often has, have would make such a variance negligible)


I'm not a complete idiot, just a partial idiot!!!


-The use of 360 vdc is from the datasheet and 365 vdc is from the 6G11 schematic I referred to . I was thinking they were pretty well the same


- However I did write 6L6 where I meant to right 6L6GC


I was actually asking;
If changing the plate voltage of a 6L6GC  from 450 vdc - 360 vdc (as in the GE chart) accounted for the difference in ELR (5600- 3800)


I wasn't referring to the original  6L6 or the 6L6G or the 6L6GB etc etc . I know the all the different types.


However I have come to realize VERY CLEARLY that it is not an easily answered question
At least to someone with my limited understanding
Thanks for trying
I do appreciate the effort

[2deaf]

Does it change the effective ELR of the 6L6Gc if you run them at 365vdc on the plates like in a vibrolux 6G11?

No.  Changing the OT to one with a different primary impedance changes the ELR.

[HotBluePlates]

I was actually asking;
If changing the plate voltage of a 6L6GC  from 450 vdc - 360 vdc (as in the GE chart) accounted for the difference in ELR (5600- 3800)

As PRR said, don't use your term of "ELR"... the better term is "optimum load resistance/impedance (for maximum power output within the tube's ratings)".

Because people see "8kΩ" often on data sheets for EL84 and 6V6, but often see "6k6Ω" or "4kΩ" in relation to 6L6's, they often assume one needs to match the OT primary impedance to the tube somehow. What we're really saying is that's not strictly true, only that the OT primary impedance matters to voltage and current swings at the tube plate.

The bigger numbers are typically seen on the smaller tubes because of their smaller power ratings; the higher impedance limits maximum current swing possible for a given supply voltage. In the end, power supply voltage & current, tube capabilities and OT primary impedance all work together to set the amp's power output capability.

This should also help answer the question you didn't ask here (but often is asked) about switching out 6L6's in place of 6V6's to get more power output. It doesn't really do much unless you also have extra power supply current available and switch to a much smaller OT primary impedance (often by changing the OT for a different unit). I think guys convince themselves they hear a little more headroom when switching, but if you set a true 6L6 amp next to their 6L6-for-6V6-swap amp, the power difference is very obvious.

[Toxophilite]

ELR is not my term
I didn't make it up
It stands for Effective Load Resistance plate to plate
It's on almost every Datasheet

[HotBluePlates]

ELR is not my term
I didn't make it up
It stands for Effective Load Resistance plate to plate
It's on almost every Datasheet

So it is! I stand corrected; my eyes always glossed over the word "effective".

Then I misinterpreted what you were saying to mean some characteristic of the tube. In that case, have the examples so far clarified why the tube is only a partial consideration for load impedance selection?