ac coupling

Help Support GroupDIY:

Analog_Fan

Well-known member
Joined
Mar 22, 2022
Messages
271
Location
Universe
thanks, i found these.

UVP1V100MDD​

5 mm x 11 mm (bigger than doesn't fit)

Standard bi-polarized series for entertainment electronics.
Compliant to the RoHS directive (2011/65/EU,(EU)2015/863).

UES1C100MDM1TD​

42 Expected 6/6/2022 the rest in 2023, as many other.

5 mm x 11 mm

Bi-polarized “nichicon MUSE” acoustic series.
Suited for audio signal circuits.
Compliant to the RoHS directive (2011/65/EU,(EU)2015/863)

:geek:, right, awesome




 

Dualflip

Electronic monk
Joined
Nov 15, 2008
Messages
1,556
Location
Mexico City, MX
Looking at the schematic, C4 is the one in reversed polarity. The rest are fine.
No, not necessarily, the polarity of C4 should be chosen depending on whether the DC offset at the output of the OpAmp is positive or negative, and in practice, it can be either one. Im not sure about that OpAmp in this circuit, but perhaps it tends to produce a negative output offset, in which case, C4 is perfectly ok as it is. The best way would be to measure the DC voltage at the output of the OpAmp, and depending on the sign, you can select the polarity. In either case, im pretty sure the voltage is very small as to cause problems in the cap.

The designer of this schematic (ESP) is not an amateur and he probably already took this into account, so for those of you who want to build this circuit, just follow the schematic instead of trying to fix what is not broken.
 

Analog_Fan

Well-known member
Joined
Mar 22, 2022
Messages
271
Location
Universe
i did a second test ...
ac coupling.jpeg

i tested:
Wima: 1nF / 1000pF fail.
Wima: 10nF fail.
Wima: 47nF fail
Wima 100nF fail
Wima 330nF could work
Wima 680nF could work
Wima 1uF works

electrolytic:
4uF7 works
10uF works

On the centuria Wima caps the amplitude only drops at high frequency, above 20kHz, but so does the VCO.
It's hard to see the difference between the 1uF Wima and the electrolytic caps with my scope.
Only a current generation digital scope could tell you and if there is a difference it would be in the range of 50mV.
 

abbey road d enfer

Well-known member
Staff member
GDIY Supporter
Moderator
Joined
Jan 22, 2008
Messages
14,501
Location
Marcelland
It is UES series.
Thanks for the link.
I must say I never used NP/BP 'lytics in my entire life.
When they're needed (recommanded?) like with the THAT1646, I use two polarized 100uF back-to-back instead of a NP/BP 10uF. Except for the real-estate issue, I never saw any disadvantage.

BTW, the Nichicon Type numbering system seems rather esoteric. AFAI can see, the capacitance unit is 0.1uF and the voltage rating is undecipherable.
I feel more an dmore challenged... :confused:
 

Dualflip

Electronic monk
Joined
Nov 15, 2008
Messages
1,556
Location
Mexico City, MX
Many (eurorack) modules sold around the world have that.

Yusynth's Roland System 100 schematic archive.

M130 VCA
M110 Full voice module

some use 100K, i use 1M.
In the M130 schematic, the only 1 M resistor I see forms a voltage divider with a 560 ohm resistor, the Thevenin equivalent resistor makes it closer to 560 ohm than 1 M, so no problem there, which is different from what you are doing, which is a 1M resistor to ground. A 100K is ok for BJT, a 1M is too much. Adding a 1M to the opamp of the schematic you posted in the first post, the 4558, can produce up to 0.8 V of offset at the output, and thats with no gain and not counting input offset current, add a mere gain of 3 and now you can have up to 2.4V offset voltage at the output, which will reduce the dynamic range, something tha can be completely avoided by using a smaller resistor.
I don't see any 1M resistors to ground in the other schematic you posted.
You can use a 1 M resistor if you have an opamp with a FET input
 
Last edited:

moamps

Well-known member
Joined
Sep 24, 2005
Messages
2,241
Location
Croatia, HR
....I must say I never used NP/BP 'lytics in my entire life.
When they're needed (recommanded?) like with the THAT1646, I use two polarized 100uF back-to-back instead of a NP/BP 10uF. Except for the real-estate issue, I never saw any disadvantage...

Same here. I started using them when I changed bipolar capacitors in some SSL boards that had this type originally installed. And I have a good friend who is an audiophile with whom Muse bipolar is highly valued, you know ;).
In general, at times I was a little confused about how to turn an electrolytic capacitor in designs with bipolar power supplies, especially when I saw DC biased back-to-back electrolytic capacitors shemes at outputs in SSL consoles. Now I don't worry too much about it. Since I started dealing with electronics at a time when we had very limited values of both resistors and capacitors, I learned to combine and manage. I just laughed when a friend asked me a few days ago what to do, 33uF is not available for the next 20 weeks .....

I feel more an dmore challenged... :confused:

Probably because of the similar feeling, I like tube designs more and more. And I don't need a 4ch MSO at 1GHz that the new guys are whining about.

P.S.
I will send you some schematics to PM that I said a long time ago that I will send to you.
 

Analog_Fan

Well-known member
Joined
Mar 22, 2022
Messages
271
Location
Universe
In the M130 schematic, the only 1 M resistor I see forms a voltage divider with a 560 ohm resistor, the Thevenin equivalent resistor makes it closer to 560 ohm than 1 M, so no problem there, which is different from what you are doing, which is a 1M resistor to ground. A 100K is ok for BJT, a 1M is too much. Adding a 1M to the opamp of the schematic you posted in the first post, the 4558, can produce up to 0.8 V of offset at the output, and thats with no gain and not counting input offset current, add a mere gain of 3 and now you can have up to 2.4V offset voltage at the output, which will reduce the dynamic range, something tha can be completely avoided by using a smaller resistor.
I don't see any 1M resistors to ground in the other schematic you posted.
You can use a 1 M resistor if you have an opamp with a FET input
like i said to: Newmarket, i forgot it.

for quad i use TL074 for dual i use Mitsubishi 5218A (i got a few tubes of them) otherwise TL072.
I assumed the 1M is good after a capacitor, it's somehow tied to gnd, the signal is centered to the next stage and doesn't consume the signal.
Never though it would offset the "next" stage.
Ok, the 100K want consume much, just touched it was safe.

i also meant the the capacitor before the Hi/Low output after the 2SK30a fet.
the 1M you talk about it to center the signal output of the BA662a, the old days of 5% resistors.

Thanks.
 

Analog_Fan

Well-known member
Joined
Mar 22, 2022
Messages
271
Location
Universe
than i place them inside the of width of the dip package of the LM13700, use the room and free up other room.

LM13700 smd.jpeg

i would free up 25,8064 mm² on each side of the LM13700, both 560R are half visible on the side of the LM13700.
than i shift many parts down and maybe ad some more functionality to a 25mm wide x 110 mm PCB.
: )
 

john12ax7

Well-known member
Joined
Oct 15, 2010
Messages
2,103
Location
California, US
And I have a good friend who is an audiophile with whom Muse bipolar is highly valued, you know ;).

Don't be too quick to discount your friend. In Cyril Bateman's capacitor study bipolar electrolytics showed lower distortion than all configurations of polarized electrolytics. Good reason to make them the preferred choice in audio paths.
 

Dualflip

Electronic monk
Joined
Nov 15, 2008
Messages
1,556
Location
Mexico City, MX
than i place them inside the of width of the dip package of the LM13700, use the room and free up other room.

View attachment 94139

i would free up 25,8064 mm² on each side of the LM13700, both 560R are half visible on the side of the LM13700.
than i shift many parts down and maybe ad some more functionality to a 25mm wide x 110 mm PCB.
: )
You don't need a 0.1% resistor to do that, a 1% resistor in a smaller package can do the same thing. Look, analyze the circuit, you have a potentiometer connected to that resistor, which makes the absolute value of the resistor non-critical since you are going to trim the pot anyway, so a 1% or 0.1% is for all purposes the same thing. Second, the temp coefficient of the pot is probably higher than 200ppm/C, so it is irrelevant if you insert a resistor of 5 ppm/C in there.

This is the last time I am going to suggest it, but I strongly recommend that you save all the money being wasted on 0.1% resistors and other useless mods, and get a used copy of Floyd's Principles of Electric Circuits and Floyd's Electronic Devices. Those are very simple books that will help you immensely, because you don't know the basics, it will not only save you time, it will save you money.
 

JohnRoberts

Well-known member
Staff member
GDIY Supporter
Moderator
Joined
Nov 30, 2006
Messages
23,278
Location
Hickory, MS
Don't be too quick to discount your friend. In Cyril Bateman's capacitor study bipolar electrolytics showed lower distortion than all configurations of polarized electrolytics. Good reason to make them the preferred choice in audio paths.
I haven't shared this for almost one year... Check out (our) Sam Groner's capacitor distortion work. Links in my old post.. here

JR
 

Analog_Fan

Well-known member
Joined
Mar 22, 2022
Messages
271
Location
Universe
You don't need a 0.1% resistor to do that, a 1% resistor in a smaller package can do the same thing. Look, analyze the circuit, you have a potentiometer connected to that resistor, which makes the absolute value of the resistor non-critical since you are going to trim the pot anyway, so a 1% or 0.1% is for all purposes the same thing. Second, the temp coefficient of the pot is probably higher than 200ppm/C, so it is irrelevant if you insert a resistor of 5 ppm/C in there.

This is the last time I am going to suggest it, but I strongly recommend that you save all the money being wasted on 0.1% resistors and other useless mods, and get a used copy of Floyd's Principles of Electric Circuits and Floyd's Electronic Devices. Those are very simple books that will help you immensely, because you don't know the basics, it will not only save you time, it will save you money.
i know the 1% works because that's the only type i use.
the 0.1% is roughly 0.44€ (single price) a piece, about 10x the amount of a 1%.
Buying 100 of them drops the price and they will get used, like the other smd components i have used and are almost to be replenished.

I have beaten more than 21.000.000 people (people with or close to having a University degree) in a ongoing hardcore coding competitions, no i dind't reach the number one spot, but well below the 10.000 rank.
 
Top