# Replacing a capacitor with one of a lower voltage in an AC/DC converter

#### Eddy67716

Joined Sep 28, 2020
21
I am going to overhaul the capacitors of my Yamaha Electone E45's power supply to extend its life. The only problems is that for the 38 volt regulator, it needs 2 55 volt 10000uf Capacitors. The store only had the same uf value in 50 volts.

I did some research and it says you needs to calculate an extra 20% of the voltage in case of surges. I got 38 * 0.2 = 7.6, 38 + 7.6 = 45.6.
50 volts is higher than 45, but is that safe or would it end up blowing a fuse, bursting one of the capacitors or worse?

Schematic for Power supply is attached below.

#### Attachments

• 1.2 MB Views: 18

#### AlbertHall

Joined Jun 4, 2014
10,574
55V is not a common capacitor voltage. If you can find one and it will fit in the available space, go for a 63V capacitor.

#### Eddy67716

Joined Sep 28, 2020
21
So 50 volts won't cut it?

#### AlbertHall

Joined Jun 4, 2014
10,574
The problem is that we don't have enough information to know whether 50V would be a safe replacement.

#### Eddy67716

Joined Sep 28, 2020
21
It is dealing with 240 volts

#### Attachments

• 204.5 KB Views: 14

#### Alec_t

Joined Sep 17, 2013
11,755
It is dealing with 240 volts
No it isn't. The capacitor will be fed by one of the transformer secondary windings.

#### Eddy67716

Joined Sep 28, 2020
21
If you mean, the secondary wiring, it is supplied 38 volts from a transformer.

#### MisterBill2

Joined Jan 23, 2018
7,325
Indeed, the capacitor is quite isolated from the 240 volts mains side of the transformer. For a 38 volt supply a 50 volt capacitor should have an adequate margin, 12 volts. So I see no problem with a 50 volt capacitor. If you can get 63 volt ones that fit, they would also be OK.

#### AlbertHall

Joined Jun 4, 2014
10,574
If that 38V is an RMS voltage then the peak rectified voltage will be 53V.

#### Eddy67716

Joined Sep 28, 2020
21
Is there a way I can precisely measure the peak with a multi-meter?

#### AlbertHall

Joined Jun 4, 2014
10,574
With no capacitor connected you can only do that if your meter has a peak measurement function - cheap ones don't.

Joined Jan 15, 2015
5,667
Is there a way I can precisely measure the peak with a multi-meter?
With a quality meter measuring the RMS value of a true sine wave multiply what the RMS value is by 1.414.
If that 38V is an RMS voltage then the peak rectified voltage will be 53V.
So Albert took the 38 volts RMS and multiplied it by 1.414 to get 53 volts or actually 53.7 volts peak.

After 50 volts the normal next high rating is 63 volts which are pretty common. I would just get a few 63 working voltage rated capacitors and be done with it as suggested. Your drawing in post #5 does not reflect the whole power supply. It does not even include the transformer secondary or rectification prior to your filter caps. You also don't mention the lead type, for example axial or radial? Here is an example of a 10,000 uF Nichicon 63 WVDC radial lead capacitor. That is just a single example of one manufacturer.

Ron

#### MisterBill2

Joined Jan 23, 2018
7,325
Look at the circuit before starting the RMD versus peak volts cry!! It is rectified DC! The label on the drawing lists the DC voltage, NOT the transformer output voltage. In addition, with a constant load current the supply voltage will be a lot less than the peak voltage. And with a nominal 38 volt level, the 50 volt capacitor will be fine. 63 will work also. But avoid 45 and 40 volt caps, that is excessively close.