EL83 OT impedance question

[Pedro Vecino]

I wanted to experiment with the EL83 / 6CK6 on a EL84 single ended circuit with 250V on plate and screen grid and 5K OT with 4/8/16 Ohms outputs.

https://frank.pocnet.net/sheets/201/6/6CK6.pdf
http://frank.yueksel.org/sheets/050/e/EL83.pdf

I see that the plate resistance in the EL84 is 40K and in the EL83, 130K.
I would be interested to know the ideal impedance relation in the output transformer. It has 2.5K, 5K and 10K taps.
Can someone throw some light?

Thanks!

[mozz]

Post deleted, i thought it was referencing the 6gk6(13.2w diss). Not familiar with the 6ck6.

[Helmholtz]

I wanted to experiment with the EL83 / 6CK6 on a EL84 single ended circuit with 250V on plate and screen grid and 5K OT with 4/8/16 Ohms outputs.

https://frank.pocnet.net/sheets/201/6/6CK6.pdf
http://frank.yueksel.org/sheets/050/e/EL83.pdf

I see that the plate resistance in the EL84 is 40K and in the EL83, 130K.
I would be interested to know the ideal impedance relation in the output transformer. It has 2.5K, 5K and 10K taps.
Can someone throw some light?

Thanks!

[ATTACH=CONFIG]49711[/ATTACH]

Hi,

plate resistance Rp and optimum load impedance (for max. clean power) are not related. The latter can be constructed from the output characteristics taking care of screen voltage, quiescent current and max. plate dissipation. From a very rough estimation I get an optimum load impedance between 6k and 8k (please look up load line construction and verify for your values) for Ia=36mA (max. allowable quiescent current for 250V@9W plate dissipation) and Vp=Vg2=250V. Max. output power will be around 3W.

[Pedro Vecino]

Thanks to both.
Mozz: I assumed you were referring to another tube.
On the same amplifier this is what I get with an EL84 and an EL83:



Comparing it with an EL84 I appreciate a slight drop in volume and more tendency towards distortion. A softer tone also. I use a 8 Ohms cabinet connected to the 8 Ohm output in the tests. In both cases (EL84 ad EL83) the volume is greater when adjusted to 4 Ohms. In the case of the EL84 it does not sound better. More balanced and musical in 8 Ohms.
At first glance it seems that the 5K tap works well with the EL83 but I do not know exactly how far I am from the ideal measurement.

For those who do not know this tube (like me ), it´s a video output tube. It was used by Rory Gallagher in his AC30. Requires adapting the socket connections.
Not long ago, JJ launched a tube called EL844 as substitute for the EL84 that undoubtedly comes from the old design of the EL83 (9 watts of maximum dissipation and open structure with the pins rearranged).

[Helmholtz]

Comparing it with an EL84 I appreciate a slight drop in volume and more tendency towards distortion. A softer tone also. I use a 8 Ohms cabinet connected to the 8 Ohm output in the tests. In both cases (EL84 ad EL83) the volume is greater when adjusted to 4 Ohms.

What do you mean with "adjusted to 4 Ohms"?

I am sceptical regarding this Rory Gallagher story. Why would someone replace EL84s with the weakly EL83s and even rewire the sockets when the original EL84s were readily available? In an AC 30 the EL84s are already misused way over specs and get way too hot (even though it is not really class A), resulting in rattling tubes after a few operating hours.

[Pedro Vecino]

What do you mean with "adjusted to 4 Ohms"?

Maintaining the 8 Ohms cabinet and selecting 4 Ohms output in the amp.

I am sceptical regarding this Rory Gallagher story. Why would someone replace EL84s with the weakly EL83s and even rewire the sockets when the original EL84s were readily available? In an AC 30 the EL84s are already misused way over specs and get way too hot (even though it is not really class A), resulting in rattling tubes after a few operating hours.

I see it possible. In an amp without master, completely dependent on the volume could be that the effect of a lower power would suit his needs. Sometimes the strangest things have a musical application.
I was in a concert that he gave in 1979. Checked shirt and Stratocaster (I did not notice the amplifier)

[Helmholtz]

Maintaining the 8 Ohms cabinet and selecting 4 Ohms output in the amp.

Well this means that the load impedance of the tube gets doubled resulting in critical screen currents and dissipation. Hope you have enough spares

I was in a concert that he gave in 1979. Checked shirt and Stratocaster (I did not notice the amplifier)

I saw him with "The Taste" in 1970 using an AC30. Same year I bought my first AC30. In later years he switched to a Fender Bassman+some other amps.

EL83s in an AC30 won't last long.

[Old Tele man]

Their respective datasheets and curves will show you what you're asking for.

[Chuck H]

Their respective datasheets and curves will show you what you're asking for.

Right. Helmholtz, if I interpret the data sheet correctly then moving to a 100% high mismatch is a better fit for the el83 tube. And it's specs look like it will take more voltage and abuse than modern el84's. So the amp is making 20W instead of 30W. Not a big deal breaker on a stage where you'll need to mic it anyway. And maybe the tubes don't roast and rattle in just a few gigs.

[Helmholtz]

Helmholtz, if I interpret the data sheet correctly then moving to a 100% high mismatch is a better fit for the el83 tube. And it's specs look like it will take more voltage and abuse than modern el84's. So the amp is making 20W instead of 30W. Not a big deal breaker on a stage where you'll need to mic it anyway. And maybe the tubes don't roast and rattle in just a few gigs.

Mismatch compared to what? In which configuration/amplifier? The max. voltages for EL84 and EL83 are the same but the EL83 only allows for 9W plate dissipation whereas the the EL84 is spec'd for 12W. The problem in the AC30 is the high plate voltage and plate dissipation. If you would run the AC30 at 250V plate voltage it would be pure Class A and everything would be fine albeit with reduced output power.

[Chuck H]

Mismatch compared to what? In which configuration/amplifier? The max. voltages for EL84 and EL83 are the same but the EL83 only allows for 9W plate dissipation whereas the the EL84 is spec'd for 12W. The problem in the AC30 is the high plate voltage and plate dissipation. If you would run the AC30 at 250V plate voltage it would be pure Class A and everything would be fine albeit with reduced output power.

Well the plate resistance has been indicated as being three times that of an el84 and the el83 will take 550V unloaded (no such claim ever made for an el84) and the same as a an el84 loaded. The comparison seemed to be in accordance to me.?.

[tubeswell]

They're both (EL83/6CK6) rated for 9W plate dissipation, 2W green diss (and gm is roughly the same @ about 10mA/V)

Assuming a plate idle voltage of 300V, for centre-bias Class A operation as a single-ended output stage:

Zout = 300/(9/300) = 10k

The OT wants to be rated for about 5W minimum I'd sayK

Therefore, assuming the screen would be at 50V below the plate, I did an easy load line for Zout=10k Vp=300V Pdiss=9W (and Vg2 = 250), indicating a bias voltage of about -6.2V with 30mA plate current (and 4mA screen current*)

* The datasheets indicate an Ip:Ig2 ratio of 7.2:1 at Vg2 = 250V

If cathode biasing, you could crank all the voltages a little higher (say to 310 plate volts). The cathode resistor would need to be: 6.2V/34mA = 182R. 180R ought to be okay. Rated for 1/2W minimum.

[Helmholtz]

Well the plate resistance has been indicated as being three times that of an el84 and the el83 will take 550V unloaded (no such claim ever made for an el84) and the same as a an el84 loaded. The comparison seemed to be in accordance to me.?.

Why should the curvature of the output characteristics (= 1/Rp) give a clue to ruggedness?

My EL 84 datasheet (Valvo/Philips, 1969) has the 550V unloaded Vp value. It is necessary because plate voltage on a transformer about doubles in demagnetization phase. With plate voltages above 275V the 550V limit is easily exceeded in backswing phase.

They're both rated for 9W plate dissipation, 2W green diss (and gm is roughly the same @ about 10mA/V)

The EL84 is rated at 12W plate dissipation, 2W screen dissipation in my datasheets.

Zout = 300/(9/300) = 10k

This simple formula ignores the actual shape of the characteristic and gives somewhat high Zout values.

EL84.pdf

[Chuck H]

It was the 550V that I took to indicate relative tolerance to electronic abuse (not ruggedness). I didn't know there were el84 data sheets with the same spec. And I hadn't considered back swing voltage, though it makes sense now. I've had to deal with that before to rate snubber parts that were burning up.

I thought that plate resistance was proportionate to load impedance.?. Still trying to understand tube charts.

[Helmholtz]

I thought that plate resistance was proportionate to load impedance?

Not at all. Rp is not a constant value. It varies with momentary plate voltage and current. Otherwise the output characteristics would be straight lines.