Germanium Amplifier

For the three and more legged things
Post Reply
brig001
Old Hand
Posts: 390
Joined: Sat Nov 13, 2010 9:56 pm
Location: Back home in Preston now

Post by brig001 » Fri Jan 21, 2011 7:28 pm

I'm not sure I follow all of your reasoning, and it's a long time since I did any of this, so I'll bow to your better judgement. I'll play with my project (it was so close to getting a case too) and update with the results.

Brian.

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Sat Jan 22, 2011 2:55 pm

Don't worry I'm often confusing :lol:

If it were me just add the two caps on the output end and that will suffice. Starts getting too complicated to do much more.


 
 
"The beer was so flat it could have been served in an envelope...."

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Sun Jan 23, 2011 12:49 pm

brig001 wrote:so I'll bow to your better judgement.
Not sure about that, most of the time I bumble around and discover things by accident :D
 
"The beer was so flat it could have been served in an envelope...."

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Thu Jun 23, 2011 1:26 pm

The trouble with designing for germanium transistors is you're always coming up against a brick wall legended (in best spray-can graffiti style) "can't do it won't go high enough".

The basic problem is germanium transistors were / are crap.

Crap max. Voltage limit Vce (too low); crap current handling capacity Ic (too low); crap gain Hfe (too low); crap bandwidth Ft (too low); crap power handling Ptot (too low); crap junction temp (too low); crap leakage current (too high).

Image

Image


I'm gonna hafta resort to extreme measures, like, use something else.


Also it's doing my head in constantly trying to think in terms of positive earth, it's just an accident waiting to happen when the wretched thing will be connected to its supply polarity the wrong way round, and that's it goodnight Vienna, so sod it it's all negative earth from now on.

Whinge #3 ~ the driver transformers are way not big enough IMO and pretty sure the freq. resp. is pretty crap as well (not enough interleaved layers – actually, no interleaved layers!), so the complete design quickly gets too complicated in various efforts to compensate, so they need to be biggerer so can drive the o/p trannies direct. Which was always the way originally. When I say 'originally' I mean back in 1960-whenever it was....


 
 
"The beer was so flat it could have been served in an envelope...."

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Mon Jul 11, 2011 8:53 pm

I think I know how I'm going to do this now.

However with two larger transformers in the box it won't be "integrated" anymore, there'll barely be room in the wood sleeve case to shoehorn in the power amp & its PSU by themselves. :shock:

Coupla phono inputs, speakers out posts; vol pots on the front and that's yer lot.
 
"The beer was so flat it could have been served in an envelope...."

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Tue Jul 12, 2011 7:24 pm

RS bits came this morning so we could be within a couple of days or so of getting the germs fired up in anger in some fashion or another.

One of three things will happen ~

A. It will work.

B. It will work but be crap.

C. We release the magic smoke.

Still a way off of getting the special transformers, so, at the mo can only concentrate on will they, won't they, bias up properly ... :shock:

Place your bets now......
 
"The beer was so flat it could have been served in an envelope...."

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Thu Oct 18, 2012 3:21 pm

Remember this? Actually been doing some work on this last couple of days:


First of all find a case of some sort ~ the original idea of using an old wood sleeve case I've got wasn't going to work, not big enough.


Metal bashing a-go-go:

Image


Image


Image


Image
Last edited by Mike H on Thu Oct 18, 2012 4:16 pm, edited 1 time in total.
 
"The beer was so flat it could have been served in an envelope...."

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Thu Oct 18, 2012 3:41 pm

Next, how to make the output stage work.

Example ~ this is the Burns Orbit 3 which somebody posted on WD forum I think. I think it's a guitar amp actually ~

Image

Note the DCR of the phase-splitter transformer secondaries are written, 2 Ohms, fortunately for me!

This doesn't look like it will work very well, but nonetheless ~ for each half of the o/p stage, the secondary winding DCR forms a potential divider with a 220R bias resistor. The very low values chosen are because of the possible issues with collector-to-base leakage current you typically get with germaniums, which tend to knacker up any ideas about more 'normal looking' biasing set-ups; this method gets around any problems in this area by merely swamping it with 'brute force'.

I was ummimg & ahrring about how to do this, and LTspice didn't help much as it doesn't seem to quite 'get' how germaniums work in reality, or not these ones anyway.

Typically, it turned out to be much, much simpler in practice.

I thought the 220R resistors a mite 'OTT' in terms of low value, so had opted for 470R.

And as it turns out, the transformers I've got have 3.5 DCR on the secondaries. More of that in a mo ~ anyhoo, try wiring it up as per LTspice and testing it with real Volts and see what happens....

Image

Thank the lord for current limited bench PSU's, it bottomed out big-time! :shock:

The LTspice model is telling fibs then.....
 
"The beer was so flat it could have been served in an envelope...."

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Thu Oct 18, 2012 3:56 pm

Ok so, I'd got biasing resistors in series with the secondaries to lift the bias Voltage (of 10R originally!), so kept reducing them until the PSU stopped limiting.

They ended up as 0.47R ~ plus DCR of 3.5, = 4 Ohms, to get the transistors biased on and conducting lowly.

Going back to the Burns, and given that the bias resistors are increased to 470R, if you divide all these values these by 2 you roughly end up with the exact same values as the Burns uses. How about that?


+Vee @ 34 Volts ~

Image



.... and here sez o/p stage is drawing 45 mA (but I think meter is under-reading slightly) [2nd scale from the top, 150 mA FSD] ....

Image

This is the resistor bias chain plus the transistors, the resistor chain is 33 - 36 mA so the transistors are both passing 12 - 15 mA -ish.


The output DC point is ~

Image


I left it idling a couple of hours and nothing changed, so looks like it might be a case of 'Arise, Sir Goer' (itsa goer)
 
"The beer was so flat it could have been served in an envelope...."

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Thu Oct 18, 2012 4:04 pm

Still working!

Image


Image



Close-up of the phase-splitter transformers, I had these made specially. 2 x 1.2k : 2 x 32R. They are well 'OTT' for the job, but originally I was going to use weedier little PCB mounting ones which I subsequently realised (by using something similar for something else) might not have a decent freq. response actually.

Image
 
"The beer was so flat it could have been served in an envelope...."

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Thu Oct 18, 2012 4:11 pm

As mentioned previously front-end is cribbed directly from 'bonkers MOSFET amp' i.e. the IT will be MOSFET driven.

Circuit ~ (link only as is large picture)

http://livinginthepast-audioweb.co.uk/i ... 2012-2.gif

Subject to change at a moment's notice of course. :D
 
"The beer was so flat it could have been served in an envelope...."

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Fri Oct 19, 2012 10:17 pm

Finished populating the board ....

Possibly .....

If it works ......


Image


Image


Image
 
"The beer was so flat it could have been served in an envelope...."

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Sat Oct 20, 2012 1:41 pm

I can tell youse all is either underwhelmed, overwhelmed, don't know what to write or aren't looking :lol:


Power supply next ~ tested the mains side wiring last night, in the process discovered the indicator lamp I put in wasn't actually a neon, but a filament bulb, survived the mains for precisely a millisecond

First need to update the schematic a bit, turns out the mains TX is 2 x 35V, + 10% overwind, not 30V as I thought

Laters
 
"The beer was so flat it could have been served in an envelope...."

User avatar
pre65
Amstrad Tower of Power
Posts: 16649
Joined: Wed Aug 22, 2007 11:13 pm
Location: Ashen Sudbury Suffolk

Post by pre65 » Sat Oct 20, 2012 1:49 pm

Still watching Mike. :)
People are like stained-glass windows. They sparkle and shine when the sun is out, but when the darkness sets in, their true beauty is revealed only if there is a light from within
Dr Elisabeth Kubler-Ross

G-Popz THE easy listening connoisseur. (Philip)

User avatar
Mike H
Amstrad Tower of Power
Posts: 17103
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

Post by Mike H » Sun Oct 21, 2012 1:13 pm

Good-oh. :thumbleft:

Big day today is testing the PSU board. (After groceries shopping, just got back.)

Came across a problem last night trying to finish it ~ the mains TX is 2 x 35VAC, OK that will work with reservoir caps of 50V....

No it won't, as it's 10% overwound so each winding is more like 39VAC off-load, which pushes the DC over 50V.

So had to scratch around a bit and resort to caps in series, means the reservoir values are half what I originally wanted.

I remember having this problem before with this transformer ~ but as each winding has a CT (interestingly), what I did then was make like 2 full-wave supplies and piggy-back one on top of the other.

It also meant the bridge rectifiers I was going to use had to be changed as well, as again 'only' 50V. Fortunately my supply of 600V ones are 4 Amps so should be OK.

Laters!
 
"The beer was so flat it could have been served in an envelope...."

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest