A guide to building the Pass F4 amplifier

[andynor]

I know you remember. You don’t forget anything! Even your own birth is probably fresh in memory. I was thinking about the other guys in here. I know I have poor judgment, but I don’t want people to believe I am bat**** crazy. Or maybe I am? I really don’t know!

 

[Chiptech]

then you either didn't read or understood what I wrote in #2623

it could save you of massive search

Zen Mod,

What I was trying to say is that I did my own research when one of the pots wasn't giving any result and from that effort I determined I had reversed the pots. I'm fine with doing my homework and this time it paid off.

But then when I still wasn't getting any juice, I reached out and Dennis offered up I had the wrong resister in R9.

But I also will admit I didn't understand Iq when you referenced it in 2623. But now I do given the postings since then.

So thanks all round.

 

[Zen Mod]

what matters is that you made it

my remark is just regarding need of patience and careful reading of replies, when question is asked - it could be case of proper reply given but not seen

I'm not picky/vane (even if Omnihighheaded) regarding many things, including effort I made in typing few lines, so no apology nor special thanks needed

 

[Chiptech]

Ok. New Problem. My left channel won't bias. I've turned P1 all the way in each direction, but no heat. P2 works.

From research it could be one of many things:
1. TL431 [post 2144 & 2161]
2. Polarity of Mosfets [post 1200]
3. Soldering of source resistors [post 1200]
4. Shorts under keratherm [post 1200]
5. try 6.8K or 5.6K at R9 instead of 10K [post 1507]
6. Try 47.5k at R8 instead of 47.5K [post 1523]

I tested the Tl431 before installing. I"m using either TL431BQLPR or TL431AIZT
I started with the Variac to test for shorts, none seen
I checked the left board against the right circuit board which biased just fine.
I checked for shorts on the Mosfets, nada

I guess I'll swap out the TL431 first, then make resistor change at R9.
Any other thoughts? Thanks

 

[Zen Mod]

from cold you ought to have same resistor values in both channels

what is voltage between N and P mosfet gates (that's A-K of TL431, convenient place for probes, take care of probes slippage) and is it varying, if you fiddle with P1?

 

[william2001]

Not saying it's the problem, but if you want the same TL431's that I use (ON Semi, 40 uA minimum cathode current), send me a PM and I can mail you some. But, assuming properly installed parts, adequate soldering, isolated mosfets, etc. You need ~4 volts from G to S to get these output devices woke up. I built my boards with a 5K at R9.

 

[Chiptech]

from cold you ought to have same resistor values in both channels

what is voltage between N and P mosfet gates (that's A-K of TL431, convenient place for probes, take care of probes slippage) and is it varying, if you fiddle with P1?

Zen Mod, I don't get the reference to A-K if TL341. Where do the probes go?

Thanks

 

[william2001]

I would check here -

 

[Chiptech]

ok, thanks

 

[Chiptech]

All Good

I started today by measuring between N - P voltage, 2.3 V still no heat.

I swiped out R9 for a 5K resistor. Still no heat.

I swapped out the TL431.

This time plenty of heat, but I couldn't control the biasing, P1 was all the way out.

So I replaced the R9 with a 10K.

This time everything worked.

I've held at 190 mV for two hours.

I'm not sure what was the issue. I did use a TL431BQLPR TL431 instead of the TL431AIZT.

So it goes.

 

[Zen Mod]

 

[Chiptech]

Here's a question:

I just noticed that my right channel biasing readings are:

Bias .176
Offset: -.001

Left channel

Bias .173
Offset: .002

Shouldn't the bias reading show negative like the offset does?

DC readings are:

Right: - 21.96

Left: 22.01

I'm looking for a misconnected wire and appreciate any thoughts.

 

[6L6]

All good readings.

Attach speakers and enjoy.

 

[Chiptech]

Several days later

After several days listening [sounds sublime] and the amp running 8 hours a day, today I noticed the right channel was louder and the balance was way off, favoring the right side - really annoying.

My measurements:
Bias Offset heatsink temp
Left 190 0 42 C

Right 190 0 52C

I lowered the right bias to 160 and the temp only lowered to 49C.

Something seems amiss - one of the Mosfets not working properly?

 

[Zen Mod]

check for uniformity:

-torque of mosfet screws
-voltage at/across all source resistors


re-check all solder joints

F4 is damn follower, if you have lower output - it's either problem with pre (swap channels test) or trivial problem with interconnects/contacts or ......... Gremlins

temperature issue - hope it'll be clearer when you check all source resistors

 

[Chiptech]

Thanks Zen Mod

I will get to work over the weekend.

 

[Chiptech]

Patience pays off.

I spent a lot of time running the amp and monitoring Bias/offset readings and heatsink temps.

Bias holds at .173/.175

Offset: near 0.001

Temp The right channel runs 4 to 8 degrees warmer over time. Is that a concern?

Sound is awesome.

 

[Zen Mod]

rotate amp for 180 deg, check temp between channels after some time

I can bet it's breeze path in room

 

[Chiptech]

Ok, I'm Stumped

Bias holding at .190

But the right channel volume is louder than the left.

Appreciate any insights on how to mitigate this?

Chip

 

[Zen Mod]

you did usual routine - swapping input cables, to be sure that problem is not upwards?

did you measured gain of amp channels in any way?