JC Fardo said:
regarding "blur 5" that's pic of Oly's little symmetric amp pcb.....not connected to Babelfish in any way,except as sideway project
,resulted as (in that moment) for me discovered "NP's way" of driving output mosfetstalking about 2SJ109BL (V) ....complement pair is 2SK389BL (V)
I have made a draft schematic of what might become my fishamp.
I don't know if it will be any good. I haven't calculated the values, but most of it is copied from chokys babbel. I have taken the CCS from aleph 30.
The most different thing about my fish is that it uses separate supplies for input and output. I have some toroids with 10x17V@1.5A and 1x18V@2A, so I figured I could try something like this.
What do you think? Is it just a bad idea?
The motivation is to let the first stage operate on a unstrained supply.
First the amp.
I don't know if it will be any good. I haven't calculated the values, but most of it is copied from chokys babbel. I have taken the CCS from aleph 30.
The most different thing about my fish is that it uses separate supplies for input and output. I have some toroids with 10x17V@1.5A and 1x18V@2A, so I figured I could try something like this.
What do you think? Is it just a bad idea?
The motivation is to let the first stage operate on a unstrained supply.
First the amp.
cviller said:
without looking at anything else.........DANGER!!!!!!!!
did you realize that entire modulation of output stage is from R211,and that resistor is referenced to -Ve,same as Q204 gate is referenced to -Ve,and that mosfet Q204 is biased also according to -Ve..........
if you split -Ve for input and output stage,you make two different references.........without further clarifications (supply modulation,supply ripple,different RC constants in two supplys).....I'll just say..........no-no
just for record............I have same thoughts.........once or twice or trice before........
cviller said:
In this thread?? 2sj109 of course!
The 2 toroids are 10 x 17V 1.5A. I've measured them with my meter and they appear to be matched with a difference less than 10mV.
ok;
check the schmtc...............look at output (tip: where are your paralleled resistors)........
don't hurry and try to understand how each block works ;
in one moment you'll hear "snap" between ears
also- you need even 10 mA through your R211 to open that gate (of Q204)........that's no-no,at least in my books
you'll fry them,sooner or later........or instantly............so-you can't use so low value for R211
Zen Mod said:
.................
also- you need even 10 mA through your R211 to open that gate (of Q204)........that's no-no,at least in my books
I meant on poor 2SJ109.....when I mention frying
Zen Mod said:
I know! Those poor exclusive fets...
I just got most of the parts I need. Still waiting for the fets though. And I'm not building anything before I'm sure I'm not just making expensive smoke!
I might just stick to the original babbelfish ccs and only change output devices. I've just read so often that people seems to prefer NP's usual ccs, but there is no point in using it if it delivers too much current.
Are the bc's better than ztx450 or only chosen due to availability? It looks like (http://www.6moons.com/audioreviews/firstwatt3/alephj.html) NP is still using ztx's. And if I'm going to use two pairs of output devices, perhaps I should also duplicate the Q4 circuit (in chokys babbel schem) as he does...
If I understand you correct; No.
You dont need two Q4 circuits, even if you use output mosfet's in pairs. Q4 turns on the CCS mosfets, no matter how many there are.
BTW Mr pass stated earlier on, that he likes to get supplies for at least 5 years at the time. Thats why you will see him allways using the same parts for many years. That probably also explanes the Zetex choice
(And IRFP240's for that matter )Steen
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.