6V6 / 6AQ5 PP Stereo Amp

Pbpix · **Posted:** Jan Tue 30, 2018 5:25 am

Flipperhome wrote:

Pbpix wrote:

Show us how it should be designed please. Why not help here a bit...

First I want to say I do NOT endorse a line powered supply, because I don't consider it safe, so purely as a thought experiment...

There's no reason for two series resistors other than to stay within power ratings and the lower 10 Ohm dissipates 5 Watt, hence the 10 Watt rating. 10 Ohm is more than enough limiting to keep max surge current under 10 A (a 1N4004 is rated 30 A). Average current is 175 mA (rounded up).

C1 has to sustain roughly 713 mA ripple. The Nichicon is rated 800 mA at 85 C and 120 Hz (The factor is .8 at 100 Hz and isn't good enough). At 132 VAC (+10 %) it has to sustain 785 mA of ripple and the Nichicon, again, makes it. Of course, a higher ripple rated cap increases reliability (the Nichicon was just the first I ran across) so the more the merrier (and costly). As a side note, when looking for 'high ripple' caps watch out for "at high frequency" types. The rating goes up with frequency and they often don't list what it is at 120 Hz (and it isn't the 'high frequency' number).

B+ current is for Class AB, 10K, which is 75 mA per PP pair (150 mA total), at a grid bias of 15 V. With plate (- cathode) voltage of 250 V that places B+ at 265 V. Simulations place B+ at 271 V but doesn't take into account AC source impedance and cap ESR losses so I figure it should come out about right, but it's close enough if not.

Edit: Corrected C1 wrong polarity.

Hi Flip:
I am resurrecting this thread here to help me understand the RIPPLE CURRENT question on capacitor selection.
( this came to light because i was seeking to better understand the ripple impact on C1 and R1.)

You indicated that the average current is approx 175ma.(rounded up)
.. ok I see that.

Can you help explain the next number?:
I don't see how you got to the next number "C1 has to sustain 713ma ripple."
"At 132 VAC (+10 %) it has to sustain 785 mA of ripple"

When I try to read up about these calculations I get a ton of complicated equations.
Did you use such equations or is there a "rule of thumb" ?
For instance I read somewhere that it is reasonable to take the load current and double it for the ripple current number.
Is that correct?

a 1/2 wave doubler is 60hz ripple right.
Ripple ratings are stated @ 120hz
and a .8 factor at 100hz

You selected a Nichicon cap with a ripple rating of 800ma

Now the next question is about the ripple adjustment factor of .8
You state that at 100hz the .8 factor ( 800x .8 = 640 ) doesn't make it.

Ok i can see that 640 drops below the average 713ma.

But then in the next line you indicate
"At 132 VAC (+10 %) it has to sustain 785 mA of ripple and the Nichicon, again, makes it."

But if the .8 factor drops down to 640ma ... how does that "make it" for the 785ma requirement?

Was that an error or a typo ... or am I missing something.

please Help ... lol
I'm getting lost.

BTW:
If I put two 220uf caps rated 800ma ripple in parallel for 440uf..... how does that effect the ripple rating?
Does it double or halve or stay the same?

Flipperhome · **Posted:** Jan Tue 30, 2018 6:43 am

Pbpix wrote:

... Can you help explain the next number?:
I don't see how you got to the next number "C1 has to sustain 713ma ripple."
"At 132 VAC (+10 %) it has to sustain 785 mA of ripple"...

I measured it in the simulation.

Even though it's half wave AFTER the diodes the ripple through (the) C1 (doubler) is full wave (120 Hz).

The 'rule of thumb' of 2x DC current for ripple might be okay if the transformer has enough winding resistance (ripple is highly dependent on the series resistance going into the first filter) but you didn't have a transformer in that design and it was hard enough to get you to put in the 10 Ohm resistor because you were touting the 'benefits' of '0 impedance' on the AC line.

Pbpix · **Posted:** Jan Tue 30, 2018 8:16 am

Flipperhome wrote:

Pbpix wrote:

... Can you help explain the next number?:
I don't see how you got to the next number "C1 has to sustain 713ma ripple."
"At 132 VAC (+10 %) it has to sustain 785 mA of ripple"...

I measured it in the simulation.

Even though it's half wave AFTER the diodes the ripple through (the) C1 (doubler) is full wave (120 Hz).

The 'rule of thumb' of 2x DC current for ripple might be okay if the transformer has enough winding resistance (ripple is highly dependent on the series resistance going into the first filter) but you didn't have a transformer in that design and it was hard enough to get you to put in the 10 Ohm resistor because you were touting the 'benefits' of '0 impedance' on the AC line.

Well ... ok.. thanks... I understand some.
But ...
If 2 times the DC current is conditionally maybe "ok" ... and if this load is 175ma ... and double that is 350ma..
that doesn't seem anywhere close enough to the 713 or 785ma you arrived at.

So.. could that work?

I do have the 5 ohm surge resistor and I do fully agree with it's effective "softening" on the 1st cap surge always.

One of the things lead to this topic today... is that... I was using the 85 degree 220uf @160v Panasonic cap w/830ma ripple rating.
But later I increased it to a 105 degree 470uf @200v (Taicon brand) it has (AQ) printed on the side so I think that's the series. However I have not been able to find this cap in their spec tables.

Anyway... I found that R1 (5.1 ohms 5watt ww cement) was "failing" open every now and then.
So that made me question if this 470uf @200v TAICON (AQ) cap had sufficient ripple rating.
(maybe you can find that ripple rating spec somehow easier than I could?)

So until I know if insufficient ripple rating is in fact part of the problem I have now instead used two of the 85 degree 220uf 830ma ripple Panasonic caps in parallel for 440uf.
I've also increased the 5 ohm surge resistor to 10watts.
The other two filter caps are each 85 degree 820uf @400v Nippon Chemi-con SMQ series rated 3.25A ripple

btw... I did read elsewhere that doubling up or parallel caps is an effective way to increase the ripple rating.... Do you concur?
REF: https://www.electronicspoint.com/thread ... nts.97409/

also from that link... this:
" Since my load for the converter is pretty constant (other than
ON/OFF), does that ease the requirement for ultra-low ESR caps?

Not at all. The ripple current is related to the value of
the output current, not how fast it changes. It is the load
current being drawn from the input capacitors, but only
during part of the converter cycle, and not at the other
part of the cycle. Remove the average from that pulse train
and you get an alternating direction current of about half
the DC output current."

Flipperhome · **Posted:** Jan Wed 31, 2018 12:47 am

Pbpix wrote:

...One of the things lead to this topic today... is that... I was using the 85 degree 220uf @160v Panasonic cap w/830ma ripple rating.
But later I increased it to a 105 degree 470uf @200v (Taicon brand) it has (AQ) printed on the side so I think that's the series. However I have not been able to find this cap in their spec tables....

https://www.aliot.com.ua/pdf/taicon_aq.pdf

Pbpix · **Posted:** Jan Wed 31, 2018 1:03 am

Flipperhome wrote:

Pbpix wrote:

...One of the things lead to this topic today... is that... I was using the 85 degree 220uf @160v Panasonic cap w/830ma ripple rating.
But later I increased it to a 105 degree 470uf @200v (Taicon brand) it has (AQ) printed on the side so I think that's the series. However I have not been able to find this cap in their spec tables....

https://www.aliot.com.ua/pdf/taicon_aq.pdf

Thanks flip... that must be an updated table that I couldn't find on their home page.
The AQ table there did not go upt to 470uf.

But.. Wow... rated 1100ma... so I guess that wasn't a contributing factor to the 50hm "open" failures .
I had 3 of them in about as many weeks... so ... well now I'm at 10watts. See how ,long it lasts.

6V6 / 6AQ5 PP Stereo Amp

Who is online