Why Are There Electrolytic Capacitors On My Inputs?

Gepetto

Administrator
Staff member
Joined
May 15, 2011
Messages
13,473
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#23
Hi Mark
When you put 2 caps in series you halve the capacitance, so it is 50uF. They are back to back to form a non-polar electrolytic (this is how they construct non-polar electrolytic caps). But they still suffer leakage current (but probably very small in value). You want a large capacitance here because you really only want to servo off the DC component of the output and this circuit is what is usually applied to do this function. It is an integrator so you want the pole at the origin or as close to that as you can realistically get it. Leakage in an integrator is not ideal. You want as perfect a cap as you can achieve.

Your 0.1uF are too small for effective low frequency response in this circuit application.

2N4401 are commonplace digital transistors. The 2N5088 would be a better low noise choice if that was what was originally specified.
 

mlucitt

Veteran and General Yakker
Joined
Jun 24, 2011
Messages
3,339
Location
Jacksonville, FL
#24
I forgot that capacitance formula, it has been a while.

I was taking a chance on the .1uF caps, but they were handy. I'll order those 68uF MLCC caps and give it a try.

Thanks for your help.
 

Gepetto

Administrator
Staff member
Joined
May 15, 2011
Messages
13,473
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#25
I forgot that capacitance formula, it has been a while.

I was taking a chance on the .1uF caps, but they were handy. I'll order those 68uF MLCC caps and give it a try.

Thanks for your help.
Hi Mark
Give the 100uF part number a try in the integrator section too. They will be lower leakage than the electrolytic capacitors that are in there now
 

mlucitt

Veteran and General Yakker
Joined
Jun 24, 2011
Messages
3,339
Location
Jacksonville, FL
#26
2N4401 are commonplace digital transistors. The 2N5088 would be a better low noise choice if that was what was originally specified.
I assume that the originally specified 2N5088's complement, which is 2N5087, is a better device than what was called out on the schematic (2N4250). Because someone before me replaced those six 2N4250's with the 2N5087's.

Give the 100uF part number a try in the integrator section too. They will be lower leakage than the electrolytic capacitors that are in there now
So, a single 100uf rather than two 100uf back to back (50uf)? Easy peasy.
 

Gepetto

Administrator
Staff member
Joined
May 15, 2011
Messages
13,473
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#27
I assume that the originally specified 2N5088's complement, which is 2N5087, is a better device than what was called out on the schematic (2N4250). Because someone before me replaced those six 2N4250's with the 2N5087's.



So, a single 100uf rather than two 100uf back to back (50uf)? Easy peasy.
Hi Mark
2N4250 are pretty much obsolete. And are inferior to 2N5087 devices.
 

orange

Veteran and General Yakker
Joined
Jul 6, 2010
Messages
17,704
Tagline
Broken beyond repair but highly affable
#28
You gotta Leggo those Eggos.
 

orange

Veteran and General Yakker
Joined
Jul 6, 2010
Messages
17,704
Tagline
Broken beyond repair but highly affable
#30
The lithium almost KILLED ME in 2012, if you're going to take a poke, be reasonable.
 

orange

Veteran and General Yakker
Joined
Jul 6, 2010
Messages
17,704
Tagline
Broken beyond repair but highly affable
#31
My blood sugar has been a bit high for a couple daze and maybe I haven't taken enough care of it. Sorry if I'm too goofy. Good night.
 
Top