Zen -> Cen -> Sen, evolution of a minimalistic IV Converter

[NicMac]

Hi Patrick,

Why use a Relay instead a simple switch?

The floating supplies need separate "switches" and relays are more convenient as you may control multiple with a single simple switch.

 

[EUVL]

Well, one can in theory also use those multi-deck switches as used in the Cordell THD Analyser.
http://www.diyaudio.com/forums/solid-state/154260-my-implementation-cordell-distortion-analyser.html




Patrick

 

[Nikon1975]

Are normal small signal relay appropriate, as they have to switch DC and signal at the same time. Any Omron series reccomended ?

 

[NicMac]

I think Patrick uses Omron G6S.
The relay only switch power (and not signal), but or course that may also affect the signal

 

[EUVL]

The signal current does not go through the relay.
Check out the current loop from DAC iout to DAC Gnd.
The battery (and hence the relay) is not in that loop, or at least should not be.




Patrick

 

[samoloko]

Patrick
does 18VAC 1.6 A transformer proper choice for charging 4 x 14 x 1.4 NiMH batteries with one LM317

 

[EUVL]

Why not make up a simple Spice file to check it for yourself ?


Patrick

 

[LuisMCP]

Patrick
does 18VAC 1.6 A transformer proper choice for charging 4 x 14 x 1.4 NiMH batteries with one LM317

I believe that each stack of 7 cells will "eat" 60 mA. This is 60 x 4 = 240 mA

18V x 0,25 A = 4,5VA and yours is 1,6A x 18V = 28,8 VA

seems OK

 

[guglielmope]

Hello Patrick,
I'm testing my SEN on ES9018 with Salas Shunt PSU.
Everything seems ok, But I'd like to have your suggestion about the shunt current. The total Idss is about 33mA and now I'm using a shunt current of about 70mA.
What do you think about?
Ciao
Guglielmo

 

[EUVL]

There is no dynamic current in the loop, so in theory 35mA is enough.
I consider 50mA more than sufficient, and 70mA a bit over the top.

But no harm done in any case.


Patrick

 

[RollE2k]

Although with salas shunt, it's recommended to shunt at least 70mA on top of load, and more seems better. So i would go for around 100mA with your SEN.

 

[EUVL]

If it should need 70mA extra to drive a 35mA load, then maybe you should consider using something else, like Walt Jung's shunt.


Patrick

 

[guglielmope]

For the moment I'm only making tests.
But I think that I'm not making an eco friendly I/V because a resistor I/V is better to save the energy of the planet.
For now the problem to solve is to obtain low noise and after I'll try other shunt configurations.
Thanks and I'll keep you informed.
Ciao
Guglielmo

 

[EUVL]

> a resistor I/V is better to save the energy of the planet

Then use batteries followed by a passive LC filter.


Patrick

 

[NicMac]

and

>For now the problem to solve is to obtain low noise

Then use batteries followed by a passive LC filter

 

[RollE2k]

If it should need 70mA extra to drive a 35mA load, then maybe you should consider using something else, like Walt Jung's shunt.

Of course you can shunt less, but as usual performance is better the more you shunt (as long as you can cool it).
But the minimum current of SSLV1.1 in original design is 100mA if i'm right, but you can always change some fets to lower it.
It's just as simple as performance vs power-usage/heat.

 

[NicMac]

> It's just as simple as performance vs power-usage/heat.

Even if this is the wrong thread I'm curios towards the validity of this postulated "simple" correlation between performance vs power usage.

1) Is it true for a constant current loads?
2) What is the correlation coefficient? You get 10% or 0.001% better performance by doubling the power usage?

curios minds want to know....maybe I should ask Salas

 

[samoloko]

when using sen Is not It better to have a lower Idss jfets In lower pair - ccs

 

[EUVL]

No.

 

[jan.didden]

Good morning Patrick from sunny Alicante - I say, you're an early riser on Sunday

jan