Merlin vs the datasheet (6AK6 SE)

Discussion in 'Shock Brother's DIY Amps' started by gigante, Feb 14, 2021.

1. giganteTele-Meister

Posts:
161
Joined:
Mar 7, 2016
Location:
Kansas City
http://www.valvewizard.co.uk/se.html
https://frank.pocnet.net/sheets/049/6/6AK6.pdf

Merlin talks about finding the transformer load impedance. He uses the rule of thumb 'Z = Va^2 / Pa'
Using 180V for the plate and 2.75W for the max dissipation gives a load of 11.78k. The datasheet recommends using 10k.
Merlin's rule would result in 165V if using 10k. How much difference would there be?

Merlin also says "the HT in a Class A amp must never be more than half the maximum peak anode voltage rating of the valve, given on the data sheet."
The datasheet shows 300V max plate, 180V for class A. So, plate voltage could swing to 360V. Is this 300V for the bias point (not peak) or more of a suggestion?

The 10k load with at 180V plate means plate voltage will max out at about 325V when biased at 98%. Maybe I just do that?

@printer2 and @shortcircuit - I know you guys like 6ak6. What have you tried and gone with?

Last edited: Feb 14, 2021
2. printer2Poster Extraordinaire

Posts:
7,196
Joined:
May 24, 2010
Location:
I went p-p. 16k roughly. Used a line transformer with a 1/8 watt tap. Not all that much in the way of help for a SE amp but I did take the plate voltage in the 230V range. I read that someone took them to 250 in a Marshall styled amp. I would go 180V with 5k on a pair. The reason the datasheet has 180V? Rectify line voltage and that is what you get. Nobody is going to use this budget tube in anything more demanding, right? After all, it was designed for portable applications, that is why the low heater draw. One thing you could be assured is that at datasheet values the tubes will last for years.

gigante likes this.
3. Ten OverTele-Holic

Posts:
738
Joined:
May 13, 2015
Location:
Central California
The data sheet doesn't give you the maximum peak plate voltage and most data sheets don't give you this information. The maximum plate voltage given on the data sheet is at idle and it is for any configuration. This isn't the maximum peak plate voltage that Valve Wizard speaks of.

180V plate voltage is just an example of a typical class A setup. The maximum plate voltage at idle is still 300V for class A.

FenderLover and gigante like this.
4. giganteTele-Meister

Posts:
161
Joined:
Mar 7, 2016
Location:
Kansas City
Thanks to both.
Didn't think of that. Nobody with any sense would do any serious design with this tube, for sure.

Perfect.

5. printer2Poster Extraordinaire

Posts:
7,196
Joined:
May 24, 2010
Location:
No, the tube probably will not like 300V at idle. It is a small tube and there is not much space between the elements. I think 250V would be the maximum in Class AB. The maximum plate dissipation is listed as 2.75W Might be able to push it to 3W (add in the screen current). So if we have 300V at idle we would need to reduce the idle current to 10 mA. Stick a dot on the plate curves graph at 300V and 10 mA, the maximum signal voltage into the grid will have to peak out at -12V to keep within the dissipation limits of the plate. You could operate at this point if you had s 30k primary winding. But again, the internal structure will probably see things arcing.

Perhaps you misspoke?

6. Ten OverTele-Holic

Posts:
738
Joined:
May 13, 2015
Location:
Central California
The data sheet doesn't mention whether the tube likes it or not, but it does say that it will tolerate it.

"It is a small tube and there is not much space between the elements."
Apparently the engineers that designed it felt the spacing was sufficient to meet the published ratings.

"I think 250V would be the maximum in Class AB."
That tube doesn't know what class its in. That tube don't care what class its in. That tube don't need no stinking class. You just can't put any more than 300V DC on its plate.

"The maximum plate dissipation is listed as 2.75W Might be able to push it to 3W (add in the screen current)."
The screen current doesn't have anything to do with the maximum plate dissipation.

"So if we have 300V at idle we would need to reduce the idle current to 10 mA."
Reduce it? From what? We could idle it at 9.1666667mA if we wanted, maybe for class A or something. We would probably be more prone to idle it at 6.416667mA for class AB.

"Stick a dot on the plate curves graph at 300V and 10 mA, the maximum signal voltage into the grid will have to peak out at -12V to keep within the dissipation limits of the plate."
What signal? We don't have no signal. We don't need no stinking signal. 300V is a DC parameter.

"You could operate at this point if you had s 30k primary winding. (sic)"
Once again, 300V is a DC parameter and you're not required to make the the maximum dissipation at the maximum plate voltage your operating point.

"But again, the internal structure will probably see things arcing."
Think so?

Perhaps you misspoke?
How? I was quoting a published data sheet. You weren't.

Posts:
8,318
Joined:
Dec 29, 2012
Location:
United States
No disrespect Ten Over because I know you know what you're talking about but Printer2 is speaking from an amazing amount of building and design experience. When he talks I listen.

8. Commodore 64Friend of Leo's

Age:
45
Posts:
4,261
Joined:
Mar 1, 2010
Location:
Kent, OH
I like it when experts argue. I get to learn.

9. printer2Poster Extraordinaire

Posts:
7,196
Joined:
May 24, 2010
Location:
I was going to reply that the OP can decide which opinions to consider, I only replied because I did not want to see him put 300V on the plates and then think he could drive it like normal in class A. I gave the mA that the tube would be biased at for 300V, it did not seem practical limiting the signal voltage to -12V to satisfy the plate dissipation limit of the tube. But if you want to design a circuit to do that, go right ahead.

I have learned a lot from a member of another forum that regularly blows up tubes to see what their limits are. He has found some datasheet parameters can be exceeded, as 6V6's are regularly, some do not 'like it'. The 12L6 and its family of other voltages. It has a maximum plate voltage of 200V and 125V for the screen. He has found he can put 360V on the plate but he said be respectful of the screen voltage, above 150V you see sparks. He says that 6AQ5's do not like much more plate dissipation than listed on their datasheet, the distance to the tube wall is not that great and any localized hot spot melts the glass. But a 6AQ5 was designed to be a budget tube, its specifications do not like to be bent all that much.

The 6AK5 was from the same era, also not a lot of excess capacity to exploit. Not a lot of people have used it other than in SE use as described in the datasheet. My opinion of the tube comes from reading what others have done and my own experience. On the tube not knowing what class it is in, if you respect the plate dissipation then your bias will be lopsided to the cold side. Basically you would have it biased like in Class AB but without the other tube to produce the rest of the signal. So you are right in a way. The tube will not care if it is biased and run that way but it might not make for a great listening experience. On the 300V limit, the datasheet says AF duty, not specifically as the output stage. Especially with guitar and massive distortion. The kickback voltages are not exactly what the designers were envisioning. I do recall someone saying they ran the tubes up to 260, possibly 280. Again, I look at it as a practical point. You need a high impedance transformer to run it here, I have used the tube at 250V and above but found no added benefit to doing so.

On the 3W for the plate and screen. It was a quick, back of the napkin example where I meant to say there is 3W across the tube, 2.75W on the plate, 0.25W on the screen. It made for neat calculations with 300V. Answering the post was already eating into my TV time, I did not plan on sharpening the pencil all that much. Why would we reduce the current "Reduce it? From what?", From the datasheet example given in Class A. When you increase the voltage across the tube you have to reduce the current otherwise you end up going over dissipation. Yes, you can run it at 300V and 9.1666667mA if you wanted to. But it would not be considered Class A. More like Class C. As far as your saying we could run in Class AB at 6.4 mA, sure. But my response was to it being run in Class A.

On the dot at 300V and 10 mA. Sure, ignore the signal if all you want to do is heat up the air and not amplify anything. On not requiring to make maximum dissipation at 300V, so you want to run the tube even colder, even more in Class C? What are you using the tube for? Are you amplifying something or not? Sure you can wire up a circuit as you say but it really would be pretty useless for our intended purpose.

On you misspeaking, I wanted to put it in a polite way that maybe your thoughts on running the tube in Class A at 300V was not what you intended. Because it is pretty darn stupid to do it. That is where this came from. I did not want the OP to think it was a good idea. You seem to think it is. I gave my reasoning why I thought it wasn't. "But it is on the datasheet" Well that is just a spec. the designer is suppose to take the specs and come up with a reasonable design. Running this tube in Class A at 300V is not a reasonable design. About all the time I will waste on this, take it as you will.

Speaking of Class C, the tube was used as a transmitter tube, graph on second page.

6AK6 used as a gain stage and phase inverter for 6L6's as a radio transmitter.

https://www.pearl-hifi.com/06_Lit_A...Sec_12/435_Single-ended_Push-pull_Amp_IRE.pdf

Last edited: Feb 15, 2021
10. printer2Poster Extraordinaire

Posts:
7,196
Joined:
May 24, 2010
Location:
Experts?

gigante likes this.
11. Ten OverTele-Holic

Posts:
738
Joined:
May 13, 2015
Location:
Central California
Not only is it not what I intended, but it is not what I said nor even implied.

There is absolutely nothing in this sentence to give one the impression that I advocate running this tube in class A at a plate-to-cathode voltage of 300V. You are refuting something that I never said.

12. printer2Poster Extraordinaire

Posts:
7,196
Joined:
May 24, 2010
Location:
And that is about all I care to say on it.

13. giganteTele-Meister

Posts:
161
Joined:
Mar 7, 2016
Location:
Kansas City
Now that it's settled I will be using 180V, is there any opinion on this part of the question?

Is this within the standard error for the 'rule of thumb'? Is this tube somewhat of an outlier from that rule? Looking at the operation characteristics, I would read the ideal load to be a little higher than 10k for peak power with lowest distortion, but that's not really what I want a tube amp to do. It's my understanding that even order harmonics are generally considered to be more pleasing than odd. So what about an 8k load instead of 10k? Is this just something I have to play with and see what happens? At what general point do I go from 'not peak output power at lowest THD' to 'tube life shortening'?

Posts:
8,318
Joined:
Dec 29, 2012
Location:
United States
I wouldn't load that tube beyond 8k because 3rd harmonic distortion begins to rise and 2nd begins to fall. In general, the 2nd order harmonic sounds more pleasing to the ear than the 3rd, 5th and 7th.

Minimum harmonic distortion in tube amps is not really a good thing, especially for a guitar amp.

Too many people confuse distortion with noise. Oh man, that is my new tag line

Theorage likes this.
15. giganteTele-Meister

Posts:
161
Joined:
Mar 7, 2016
Location:
Kansas City
Rob, I think we're on the same page here.

robrob likes this.
16. giganteTele-Meister

Posts:
161
Joined:
Mar 7, 2016
Location:
Kansas City
So, this means the best choice for my OT is probably the 1760C. I might even put a switch do go from the 5k to 8k winding.

That would give me higher power, clean at 3.2, 8 or 16 Ohm on one and lower, more 2nd order at 1.6, 4 or 8 Ohm. There used to be a guy that posted here a lot that had something about a switch in his tag line....

I don't have any use for the 3.2 or 1.6, so I'd probably cap that one off.

17. printer2Poster Extraordinaire

Posts:
7,196
Joined:
May 24, 2010
Location:
Last edited: Feb 16, 2021
18. giganteTele-Meister

Posts:
161
Joined:
Mar 7, 2016
Location:
Kansas City
Yes, forgot I have two separate threads going that seem to be converging. Parallel SE. I did see that article but only now went back to it because after I read the amp was changed to PP, I stopped reading. I also considered an option to pull a tube, but that approach is cool.

It will do PP if you change the load from resistive to reactive.

19. printer2Poster Extraordinaire

Posts:
7,196
Joined:
May 24, 2010
Location:
Yeah, I am not immune to user error at times. Spent the last two hours playing around with the graph and the different tubes I have. I was suppose to be working on my guitar, darn.

20. giganteTele-Meister

Posts:
161
Joined:
Mar 7, 2016
Location:
Kansas City
I was supposed to be working. Full stop. Ooops.

IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.