F6 Amplifier

[Zen Mod]

you are saying that all your F6 will not be the same ?

or that all ours* F6's will not be the same ?



-Tea , Coffee ?

-Yes , please .....

The Coca Cola Kid - YouTube

* not mine , Bubba

 

[Nelson Pass]

I meant each different design...

 

[Zen Mod]

yup

I knew that was Yes, please....... sort of answer

 

[buzzforb]

With F6 and F5T trimed to similar distortion ratio at 1K, sou d is definitely unique to each design. I hate that SS fets dried up. This amp was going to be a great teaching tool for vlamp builders. F5 is trimmable, but doesnt seem to offer the range.

 

[Antoinel]

With F6 and F5T trimed to similar distortion ratio at 1K, sou d is definitely unique to each design. I hate that SS fets dried up. This amp was going to be a great teaching tool for vlamp builders. F5 is trimmable, but doesnt seem to offer the range.

Suppose H2 and H3 are independent variables. There maybe a handle or a tweak [e.g. a Zen Pot] to increase and decrease %H2, and a second but different handle [?] which does not affect %H2; and also increases and decreases %H3 at will. Now I have the making of a two level factorial design leading to 5 experiments: A % THD analyzer is needed.

• H2 High and H3 High
• H2 High and H3 Low
• H2 Low and H3 High
• H2 Low and H3 Low
• H2 Center [(H+L)/2] and H3 Center.

The above experiments are expected to yield 5 different sonic signatures because each distortion setting is a unique combination.

So, does the design of F6 have two different "tweaking handles" to independently vary H2 and H3?

 

[buzzforb]

YEs

 

[Antoinel]

YEs

Thanks buzzforb, your answer is great news. Please specify which handle affects H2 only, and the other one which affects H3 only.
Best regards

 

[buzzforb]

Havent defined it that well but if you look back through the thread, hints are evrywhere.

 

[Antoinel]

Havent defined it that well but if you look back through the thread, hints are evrywhere.

Thanks buzzforb. The article on F6 by Mr. Pass has a wealth of info in a few paragraphs. I'll sort it out.

 

[Antoinel]

Thanks buzzforb. The article on F6 by Mr. Pass has a wealth of info in a few paragraphs. I'll sort it out.

Mr Pass's article in 6moons has a great explanation on the specific handle [degeneration and JFET matching] to manage H2 and its different manifestations. In lhquam's post #s 3184 and 3186 he separated H2, and H3 from Total %THD. For example, in post 3186 he compared simiulation with actual results on TEASER-6. At 500 Hz, He found H2 was ~0.008%, and H3 was ~0.005%.

Truly low H2 and H3 numbers for TEASER-6. Suppose one operated diyF6 in Class AB mode. Will its comparative H2 and H3 be much higher? Will more H2 and/or H3 translate to better sonics.

 

[Zen Mod]

nothing is stopping you from testing it

 

[Tea-Bag]

I completed F6 with IRF240's.
I used 1R Source resistors. The bias only went to ~500mv with existing 5K Pot.
Assuming I go, to .5R resistor, I think this will give me enough room to bias higher. I was too scared to add pots across the source resistor. With IRF240's I am not sure if that's a good idea, assuming the overall source resistance goes below <.5R overall.
If not I was considering changing R3-4.
Any thoughts?

 

[Zen Mod]

LED voltage is culprit

what you're using - blue ones , how many in series ?

and - what voltage you're getting across them ?

 

[Antoinel]

nothing is stopping you from testing it

Thanks to Tea-Bag for his high caliber supplies; R100A and the upcoming PCBs. I am slowly following your recommendation.

 

[Tea-Bag]

LED voltage is culprit

what you're using - blue ones , how many in series ?

and - what voltage you're getting across them ?

ZM - did not even think of that! The only thing Papa asks us to measure
It's 4.05v using one blue and one red Led. I will try two blue ones.

 

[Tea-Bag]

Special thanks to Buzzforb , who's Onetics transformer is in action here.

 

[Zen Mod]

next time you put something funny in tea bag , call to share

 

[Antoinel]

nothing is stopping you from testing it

Zen Mod. Here is what astounds me by using this example:

• 0.1% H2 is 1 sine cycle of H2 in 1000 sine cycles of the fundamental H1.
• The amplitude of one H1 in a reference 2.828 Vrms into 8 Ohms [or 1 W] is equal to (2.828 /1000) = 0.00283 Vrms.
• If the amplitude of one H2 cycle is equal to that of one H1 cycle, then the answer is 0.00283Vrms across the same 8 Ohms simultaneously stimulated by H1.
• What is the output power attendant to H2?
• P{W} = [V X V]/8 = 1 microWatt.
• But fortunately; human hearing is 1000 times more sensitive in detecting H2 than H1 [just playing with working numbers].
• Thus, the actual output power from H2 is 1 milliWatt.

In my previous post [Class A/B diyF6], I wanted to raise the power level of H2 to 10-100 milliWattsi in 1W total output power. Afterall, H2 is a desirable distortion; within limits.

 

[Tea-Bag]

ZM - did not even think of that! The only thing Papa asks us to measure
It's 4.05v using one blue and one red Led. I will try two blue ones.

Two Blues give room to bias up much higher. Measures closer to 4.25v now.
I just used two random IRF240's, no matching tried. Offset still good. I will try some matched Fairchilds in the other channel.

 

[lhquam]

Good looking boards. Yup - you need to put LEDs in series or use zener diodes up around 6 volts.

I completed F6 with IRF240's.
I used 1R Source resistors. The bias only went to ~500mv with existing 5K Pot.
Assuming I go, to .5R resistor, I think this will give me enough room to bias higher. I was too scared to add pots across the source resistor. With IRF240's I am not sure if that's a good idea, assuming the overall source resistance goes below <.5R overall.
If not I was considering changing R3-4.
Any thoughts?