|Solid State Talk all about solid state amplification.|
Please consider donating to help us continue to serve you.
Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
||Thread Tools||Search this Thread|
|30th August 2002, 10:38 PM||#1|
Join Date: Aug 2001
Location: Bath, UK
Matching bipolars on Hiraga 20W
A question: I've recently built a Hiraga Le Classe A amplifier and am very happy with the result. There's some pics here:
Anyway, I've lived with the amp for a few months and I thought I'd try some tweaking. I made no attempt to match the transistors when I originally built the amp so this seemed a good place to start....
I matched the input and driver stages to within 5% easily but the output stage is much harder. I used ONSemi MJ21193/94 for the output and these transistors cost about 5UKP = $8 each so I only bought two pairs and they don't match at all well.
The averaged gain figures are:
Ib = 5mA: NPN = 49, PNP = 84
Ib = 10mA : NPN=55, PNP=82
Ib = 15mA: NPN = 56, PNP = 80
Ib = 20mA: NPN = 55, PNP = 78
This trend follows through to saturation.
My question is that whether this is okay or should I buy more and try to match better. I'm ~very~ happy with the result but I wonder what else could be acheived with a little more effort. The curse of DIY, I guess.....
Any help is much appreciated.
|31st August 2002, 08:14 PM||#2|
Join Date: Oct 2001
Location: Munich, Germany
Setting bias in Hiraga power amp
the 500 Ohm pot is only for the DC offset, not for the bias current. The latter is defined by the resistors around the driver transistor.
If you reduce the 1k resistor going from base of driver transistor to rail to 900 Ohm, you will have a decrease in bias current.
You could change the 0,33 Ohms resistor from emitter of driver transistor to rail to 0,47 Ohms, alternatively.
If you go below 1amp bias current on a Hiraga design, you lose subjective power.
I also suggest redesigning/recalculating the resistor values for the input transistor to have them run at 600uAmps, as this is the optimal working point for 2SC1775/2SA872, soundwise.
You could also change the 33k resistors in the collectors of the first input transistors to 47k, that would change all bias currents through all components.
Re the choice of output bipolars: when repairing a burnt original Hiraga 30W lately, I used the Toshiba 2SC5200 and its complementary, which are the modern versions of the AKSA amp outputs. The owner of said Hiraga wasn't disappointed.
|1st September 2002, 11:27 AM||#3|
Join Date: Apr 2001
Location: Moonee Ponds, Vic, Australia
The 2SB688/2SD718 I have used are £1.30 (Ex VAT) at <a href="http://www.elclondon.co.uk/acatalog/ELC_Online_Catalogue_MORE_ON_PAGE2_136.html">East London Components</a>, they are TO3P devices though.
|Thread Tools||Search this Thread|
|Thread||Thread Starter||Forum||Replies||Last Post|
|Matching output bipolars||primare||Solid State||1||6th April 2006 08:29 PM|
|N-Channel with bipolars?||TroelsM||Solid State||10||6th March 2006 12:55 AM|
|bosoz with bipolars||rmgvs||Pass Labs||2||21st February 2006 08:18 PM|
|Darlington Bipolars as switches||rtarbell||Parts||3||5th February 2006 07:47 PM|
|Cascode cct minimises SOAR requirements of output bipolars.||Circlotron||Solid State||10||12th February 2003 12:13 PM|
|New To Site?||Need Help?|