Basic

[bcmbob]

Originally Posted by KatieandDad

This little project is EXCELLENT and not fantastically expensive.

Jean Hiraga's Super Class-A Amplifier

I looked at that one last night. Will definitely start something in the near future. Thanks again.

[KatieandDad]

If I remember correctly. In order to add efficiency the speaker transport needs to be looser. That adds to distortion.

I had a mate who used to boast that his woofers could be seen to be moving. If I run my speakers at full volume I can feel them moving but I can't see them moving.

Takes me back to the early days of Far Eastern speakers. We like rock and our speakers like rock. Makes them a bit stiffer but that's what Brit amps were designed for. Far Eastern music used to be "Twing Twang" music and their early speakers were designed for that. These days we all listen to the same music and speakers are being produced accordingly.

[KatieandDad]

Funny but I did some measurements the other week.

With the massive Aleph 4 running at "LOUD" it was only producing about 6W.

[nigel pearson]

Originally Posted by KatieandDad

Nigel, I have seen some variations of the JLH and Hiraga Class A amplifiers but I've never seen anything so different to the diagram that you have posted.

This is the classic Hiraga 8W - Unfortunately most of the components are now obsolete.

It started with the Hiraga , my friend Martina knows him and I have followed him for years . I wish I had found my finished circuit . It uses what looks like an undersized bootstrap to replace the JFET CCS . Undersized in so much that it was 10 uF where as text book would be typically 100 uF ( I reduced the size until crossover distortion appeared then went up a bit , I have a big box of 10 uF polyester , ideal ) . It also had a very cheap power supply based on a voltage doubler for the driver stage ( x 2 , +/- ) . The idea being that the MOS FET's would be driven hard enough for more class AB power . It seemed it was there for the asking and a free lunch . I didn't like it . What I did do is convert it to a clean supply of 1 V below that of the output MOSFET's . Funny thing is I have told many people this and at best they seem bored ! To have a cheap clean driver stage seems less than boring and the idea of a soft clipping option seems to me worth thinking about . The love of all MOS FET amps might be to do with this as much as anything else ? Complimentary feedback stages with bootstrap CCS need care . They produce nasty distortions at clipping if a series resistor is not included . So the assumption of < 100 % positive feedback for bootstraps perhaps should be < 95% . That makes a Darlington ideal . MOSFET's are said to be unsuitable for bootstrap CCS ( 78% voltage gain ) , not so .


Apologies for getting into circuit talk . BTW I do hundreds of " what if "circuits . This was just one . I never tried it with Quad speakers . I think it would have been ideal . Someone said they didn't recognize the simulation program I use . Easy , I don't . If you are tempted to try the circuit use an output capacitor until sure you have the DC offset right . It should be very stable once nailed . The MOS FET bias will be on an individual basis , use a 1 K pot at first . I do suspect less bias could be better . I have in mind that 4.5 W is ideal class A performance . Even reducing to 100 mA will sound good . The amp is able to offer good class B performance . 100 mA is a reference point for many Audio MOSFET's . It marks the positive temperature coefficient tipping point . The original Hitachi MOS FET circuit of 1978 is superb , no kits available as far as I know . I based my 160 VA PSU theory on that ( +/- 55V at 160 VA minimum = 90 W sustained 150 W 4 0hms transient ) .