Schematic questions

[uki]

Hey guys!

Please I would like to know few things on the following schematic:

1- What is R13 for ?

2 - What is C11 for ?

3 - What is the connection from artificial heathers center tab to V2 cathode?

Thanks in advance!

[shooter]

look up "Grid leak, Grid stop resistors"
the CT for heaters to the cathode "elevates" the heaters, used for noise reduction


believe the cap is for high frequency spikes, "transient suppression" but might wait for smarter than me before surgery 

[Jonas]

R13 is your ground reference for the power tube.
R1 and R2 connection to V2 pin 2 is an artificial center tap for the heater winding to reduce hum (the PT in your schematic does not show a center tap wire but you can "artificially" create one with the (2) 100 ohm resistors to ground. Grounding the (2) 100 ohm resistors to the power tube cathode gives you a slight DC elevation from ground which helps to reduce hum further (dc elevated)

[Latole]

believe the cap is for high frequency spikes, "transient suppression" but might wait for smarter than me before surgery 

Right , a spike protection ; 1 kv and more cap

[uki]

Thanks guys ! 

[bmccowan]

I had to chuckle. Shooter tried the old "teach a man to fish" thing but a couple folks threw Uki a salmon.
Anyways, reading up on grid leak and grid stop resistors is a good idea. Until I did, I always needed someone to throw me a fish. And now I need to go read up on spike protection caps.

[kagliostro]

Ask to ChatGpt for Zobel Network on the primary of an OT of a Tube Amp, then ask the difference if the resistor is not present (like in the schematic that was posted by UKI)

Franco

[shooter]

"teach a man to fish"

Ask to ChatGpt

The dark ages in 3, 2, 1.....

 

[uki]

Does C11 have anything to do with the Pentode/Triode switch ?

[acheld]

Does C11 have anything to do with the Pentode/Triode switch ?

Probably, maybe, I guess.

However -- and I say this with no practical experience owning and using an amp like this --  I would not recommend flipping that switch while the circuit is live.  I can imagine awful transients doing this.  C11 might tame some of that, but I can't imagine that it take care of all (as it might with a simple on-off switch). 

Hopefully, folks with real experience can chime in.

[Willabe]

However -- and I say this with no practical experience owning and using an amp like this --  I would not recommend flipping that switch while the circuit is live.  I can imagine awful transients doing this.

Yes, Do NOT flip the pentode/triode switch with the amp on. Turn the amp off before you flip that switch. 

[Lectroid]

"teach a man to fish"

Ask to ChatGpt

The dark ages in 3, 2, 1.....

 

 

Re:  Free Fish vs. Fishing Lessons.
Long ago, I read something interesting:  In order to ask an intelligent question on any given topic, one has to know most of the answer already.  I like that idea so before I compose a question here, I try to understand as much of the problem as I can--to read what the books say, work out the Ohm's Law ramifications, etc., to try to know as much of the answer as I can, before I ask.  I want to be able to ask the narrowest question I can.

But: when I ask that question here--hoping for a fishing lesson--sometimes someone just--bang!--throws me a fat albacore, right into my boat.  Well, okay, bro--thanks for all the fish.  But honestly, I would have preferred a teaching moment from one of the moderators. My point is: a) as posters we have different styles of absorbing teaching and b) we have no control over the form of the answers we get or how an answer is going to be presented to us--fish or fishing lesson?  We can only ask the question and take what we get.  It's just random chance that determines whether a fish or a fishing lesson pops out. 

That said, I fully admit it.  People here have tossed me plenty of fish and it's smoothed my path a lot at critical moments.  It's a fine line for the moderators to walk: how far to talk down to someone?  Does the OP just really need a fish right now, or instead is it lesson time?  I think the moderators here do a pretty good job in matching the style of their answers to the relative skill/knowledge levels of the individual posters.  That's been my experience anyway.

[jbrrrrr]

The dark ages in 3, 2, 1.....

 

Soon, questions here will look like:

"How do I write a prompt for ChatGPT to add three gain stages to a Blues Junior?"

[jbrrrrr]

But: when I ask that question here--hoping for a fishing lesson--sometimes someone just--bang!--throws me a fat albacore, right into my boat.  Well, okay, bro--thanks for all the fish.  But honestly, I would have preferred a teaching moment from one of the moderators. My point is: a) as posters we have different styles of absorbing teaching and b) we have no control over the form of the answers we get or how an answer is going to be presented to us--fish or fishing lesson?  We can only ask the question and take what we get.  It's just random chance that determines whether a fish or a fishing lesson pops out. 

THIS is where I find AI to be of a lot of practical value.  I'll learn as much as I can about whatever it is I'm trying to do, I may ask a question and get the correct answer from someone but not much information on the how or why - taking this kind of thing over to ChatGPT or Claude to start a dialogue and have it feed me information in ways that I can best understand it is really useful for comprehending new concepts quickly.

[uki]

Yes, Do NOT flip the pentode/triode switch with the amp on. Turn the amp off before you flip that switch.

I don't have this amplifier, the question(s) was only theoretical,
BUT i was wondering about this switch as well, the answer came without I even ask 
I bet on such amps there is no warning

I'm totally in favor of: teach how to fish

[Lectroid]

We can only ask the question and take what we get.  It's just random chance that determines whether a fish or a fishing lesson pops out. 

THIS is where I find AI to be of a lot of practical value.  I'll learn as much as I can about whatever it is I'm trying to do, I may ask a question and get the correct answer from someone but not much information on the how or why - taking this kind of thing over to ChatGPT or Claude to start a dialogue and have it feed me information in ways that I can best understand it is really useful for comprehending new concepts quickly.

@jbrrrrr:
I admit I have used AI many times to delve into many subjects where I needed a quick overview. In those cases, I've found large-language models to be a good tool to generate a summary of a given subject area.  I find such summaries to be good jumping off points for me to then turn to books or other (human-generated) online sources that I know to be trustworthy and accurate.

That said,  I still have big reservations about using LLMs for in-depth scientific, engineering, or technical information.  I don't question that it will give us some kind of an answer every time.  But is it be the correct answer, every time?  If it isn't accurate, how do you determine that?  That's what remains to be seen, in my view.  I just don't have the trust level yet. 

Can you give me examples of the sorts of questions you've asked an AI?  How would rate the level of the AI's accuracy in answers you've received?  What about depth?  Does it provide a deeper level of information compared to what you can get from, say, Rob Robinette or this place or TAG for example?  No judgement here, I'm just being curious about what you've seen using an AI.

[shooter]

questions you've asked an AI?

while Never have or never will knowingly, try;
"If you're artificially smart, how do I know your answers aren't artificial?"
or "Have you ever played with a quiji(sp) board?


 

[Lectroid]

I've read too much sci-fi to be a knee-jerk, carbon-better-than-silicon bigot.  AI presents us with immense opportunities.  But yeah...  How do I assign some level of trust to its answers? 

It feels to me like too many folks are ready to accept anything they hear it came from an 'AI'.  I don't like to see it; I wish more people could be more skeptical.  But I'm a born skeptic so I'm biased.

But being skeptical, I can't help but ask: Er...just how accurate are they?  Who's checked them for accuracy?  Whose check their math for their level of precision? 

I hope somebody's working on this.  The idea of accuracy is central to the whole 'use them or not' question, it seems to me.   Are they 100% accurate and unbiased data brokers?  What percentage of confidence in AI answers can one conservatively expect?  Just give me a number--    99.8%?   Or closer to 13.4%, on a good day?

I wish I had all these answers.  Wait--I can just go ask an AI.  I'm saved!

And way off topic.

"Open the pod bay door, HAL"

[shooter]

And way off topic.

in a round-a-bought way it's exactly on "topic"
Uki is trying to "understand" some else's design
the design Engineer isn't present so the next best thing is "experts"


once I'm "accepted" as an expert I could lead Uki down some amazingly false premises.  The only way He'd know, other "trusted experts" would call BS.


once AI is an accepted "expert" there's not enough Analog experts left to call BS

[uki]

AI  ?

I can think for my self !!

Uki is trying to "understand" some else's design

Yes !

I plan to make another amp, with the parts from a Geloso 268 reel to reel, not using the schematic above thou, but just learning if something from it worth implementing, like the connection of the artificial center tap.

[HotBluePlates]

Does C11 have anything to do with the Pentode/Triode switch ?

RDH4 Link

No, the Pentode/Triode switch has nothing to do with nothing, except being a kinda-lousy way to "reduce power output" (cause it doesn't reduce by "enough" and changes the sound to be less-good to many ears).

Ol' Langford-Smith compiled some knowledge about single-ended pentodes and their frequency-response due to the non-constant impedance of a loudspeaker on Page 566.  Right away on Page 567 he tells how the Old Designers squashed that issue (at least for high frequency).

On Page 880, we are reminded that the loudspeaker's power output is equal to the square of the signal current times the load resistance (which is changing, and rising as frequency rises).

Page 882 tells us something about how output pentodes with negative feedback handle the non-constant loudspeaker impedance.
Does your schematic show a feedback loop around the power section?

Page 881 talks about some approaches to cope if feedback is not used.

[jbrrrrr]

@jbrrrrr:
Can you give me examples of the sorts of questions you've asked an AI?  How would rate the level of the AI's accuracy in answers you've received?  What about depth?  Does it provide a deeper level of information compared to what you can get from, say, Rob Robinette or this place or TAG for example?  No judgement here, I'm just being curious about what you've seen using an AI.

I had a very long response to this typed and ready to go, but I think we've derailed this particular thread enough re: AI off of the initial joke that was made - I'll message you directly - if there's interest from anyone else, feel free to make a thread and I'd be happy to contribute my experiences using AI as a technical coach and sounding board.