P-101 Biases into Class A

[john65b]

Hey all, Before I started my KSA-50 Clone Project, I thought I would finish another project that I have wanted to do for a long time...an ESP- P101 build. I have the amp modules built as the attachment shows - +/-51V rails, 2 pairs output devices, R39 resistors, 4 ohm speakers.

Now I completed the amp a few days ago and have been tuning the BIAS as the instructions say - the target was to get around 100ma. I remember the Class A calc I downloaded from the Krell Clone Wiki (from K-Amps?) and played around with the values to maximize the current bias to maximum dissipation to my sinks, which according to manufacturer, for 4 output devices max per sink is 115W. To dissipate the 115W I could crank the bias to 490ma. Now understand I still have the originally intended power supply that the instructions called for - six 4800uf caps so 14,400uf/rail, not the usual 100,000uf/rail that I normally see on these class A amps. This under capacitance is most likely why I hear a little hum at this bias level. So I reduced the bias until the hum was gone and it is at 250ma. At 250ma I am supposedly getting 2W Class A and 55 Watts dissipated through sinks, and no hum. OK.

Anyway allow me to digress for a moment...I always thought I needed a ton of power to drive my Maggie MGIIIa speakers (4 ohm, 86dB/watt), so I built myself a UCD400 amp (never had it over 12:00 on Volume knob). But after putting a voltmeter across the output and listening at my listening position and to what I consider a loud volume, I was only getting max about 16V. At 16V on 4 ohm speakers, is really only 65 watts. I back calculated the 2 watts Class A and got 2.9V, so I reduced the volume till I was maxing around 3V on the outputs and noticed it was around the normal level I usually listen to.

Anyway I am thinking of stripping the 40V-0-40V tranny to a 22V-0-22V tranny, putting in the higher capacitance (I have four 56,000uF caps) to be able to get more Class A watts (at the expense of giving up the higher Class AB power I have at higher rails) from the P101...

Question is do I have an error somewhere in my calcs, or has anyone set their P-101 cranked exclusively into Class A? Is running the P-101 in 20 - 25 watts Class A worth it or advised?

BTW - my listening room is 22' x 12'

[MEGA-amp]

The P101 is a great amp. I built it, and cranked the bias to around 500mA and it definitely sounded better with the higher bias, granted I also had 392,000 uF total capacitance and a 1500VA transformer. My trafo secondaries were at 38V and after rectification and caps it was around 53V. Since you dont have very sensitive speakers, you wont be in class A very long when you want to turn the volume up. I do believe Rod himself stated somewhere that running the P101 in class A is not recommended as it wasnt designed for it. I dont think you need to go as low as +/-22, you could probably get away with +/-35V. If you have the proper sinks then go for it, and of course a large trafo always helps. I would use the four 56mfd caps in conjunction with your other caps and you should be fine.

-john

edit: make sure you use some sort of speaker protection

[john65b]

At 500ma, 8 ohms, +/-53V you were in Class A for 16 watts, correct?

Maybe I will try as you say - I can put the 400VA 22V-0-22V tranny, add two 56,000uf Caps, and bias up again.

My sinks would be maxed out at 700ma too (100 watts dissipated), but the 2K pot will need to be upped to 5K to get to 300mv....

[MEGA-amp]

sorry I can't really help you with the calculations. If your plan is to go class A, I wouldn't suggest it with the P101. If your 400VA trafo is handling both channels I definitely wouldn't do it, its too small. I would add as much capacitance as possible and find the largest trafo you can afford. For the most part I ran the P101 around 300mA, and I was very happy with the results. It was most likely due to the fact that my PS was built with class A in mind.

-john

[jacco vermeulen]

Not counting egg friers like MF, a lot of Class A amp models have close to equal numbers for dT(sink-ambient) and dT(junction-sink)
At the price level of the Renesas/Hitachi devices it's cost-efficient, often the highest dissipation bang for the buck.

Suppose you smuggle a little and take a temperature rise between sink and ambient of 25Cs and for the rise of junction to sink 35Cs.(T-ambient at 25C)
The best you can get for the plastic devices is some 20 watts dissipation for 35 degree temp rise, 80 watts total for a 4 output device P101 board.

If your heatsink does an Rsa of 0.30, then 80 watts dissipation puts the temperature rise of the heatsink at 24Cs.
25C ambient + 24C for the heatsink + 35C for the junction to sink= 84C junction temperature.

A 22VAC transformer allows 1.27 amp bias for 80 watts total dissipation per channel, some 25 watts Class A operation.

Roddie's P101 boards are not layed out very heat efficient with the output devices grouped so close together, a bummer for Class A operation.
Secondly, the rail voltage loss for the lateral mosfets is rather high, 22Vac won't even give you 35 watts/8 continuously.
The general practice for Class A operation of these output devices would be to use a separate rail for the front end at a higher voltage level, in order to raise the output to dissipation efficiency.
Something that requires brain surgery on Rod's PCBs (afair the P101 board layout)
Using these boards in full Class A is what we call "as a flag on a pig boat " overhere, a shame of all the expensive hardware.
(terminate me, i can't help it, double dutch heritage)

[john65b]

Will do. I will leave the P101 as is (currently at 250ma). The 80VCT, 400VA Transformer runs a bit hot tho...

I will continue to build the full blown Krell Clone (112,000uF/rail, +/-40V rails and 1240VA)

Jacco - we have a similar saying "Putting Lipstick on a Pig"...

The P101 is hardly a Pig (I know that is not what you were trying to say), and may just replace my UCD400 in the setup...the Krell will become the new challenger to the throne...

Does this ever end?

[MEGA-amp]

Does this ever end?

No, it doesn't. And it's certainly a good thing that it doesn't!!