xformer o/p w/ cap 4

[angryman]

Hello....most likely a rudimentary stupid question I'm just not getting....
Got about a 10v (DC) increase on the output of a rectifier/transformer when I put a cap across it...
Prim 110 I think
Sec. 24vac to rect. dc

how could this be? I believe the cap is rated for 40V.

 

[nbucska]

DVM-s in DC mode usually measure average
The cap charges to peak VTG.

Plesae read FAQ240-1032
My WEB: <

http://geocities.com/nbucska/>

 

[VEBill]

Or more generally:

Using meters (especially digital) is often like looking at the world thru a soda straw. Use a 'scope to see the whole picture.

 

[itsmoked]

Or again it's your meter!

Think about it.. What's coming out of your transformer/ rectifier? A bunch of humpy pulses.. Just what will your specific meter read those as?? You put a cap there and suddenly you have have tiny humps on a DC level. You will get a large increase. Your meter is still going to be wrong. Mine wouldn't be but yours will. You need a true RMS meter or work around it.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[OperaHouse]

What about the obvious! You have 24V RMS AC sine wave. Rectify that and you have pulsed DC that averages 24V. Adding a capacitor (and under light load) the voltage charges up to peak value which is 1.4 times the RMS AC value. For 24V that is about a 10V increase. The size of the cap will determine how much load you can put on it before the voltage drops back down to about the RMS value.

 

[angryman]

Wow, thanks you guys, I told you it was a stupid question, but I don't have the thing in front of me, although I should have known anyways...

thanks again.....

Mark

 

[angryman]

In an unrelated circuit, I couldn't find anything doing a search here....but how would I calculate the cap size to use to increase my voltage across the rectifier on the 24vac sec. transformer to 24v (the drop on the rectifier is causing me not to get close enough to the 24v I need..... ~22v isn't working)

thanks,
Mark

 

[nbucska]

The basic equation is

i=C*dU/dT Where i=current [A], C=cap [Farad]
dU= voltage change [V] and dT = time of the change [sec]

If you have 60 Hz with full wave rectification then
dT= 1/120 = 8.333 millisec. If you know your current and
the max. voltage drop, you can calculate it using the Eq.


Plesae read FAQ240-1032
My WEB: <

http://geocities.com/nbucska/>

 

[itsmoked]

angryman; Sorry but a cap will only do so much. You cannot just add more until the voltage reaches some value. Chances are if you have any cap now adding more won't help you. Besides adding too much can have harmful effects.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[angryman]

this particular instance is only for my testing junk...I put a 1000uF cap across the DC on the rectifier and got 32V, which according to the formula above is roughly 2.9A. A little high but for now it will do...however, I have a low accuracy meter that measured only 1A...I'm having trouble getting my Fluke 123 scopemeter to measure current properly....not sure....anyone out there got one of these things? Only thing I can figure is that the oterh fluke (low accuracy) is not accurate and the scope probe is 10x or som'...it's measuring about "32" amps!

 

[itsmoked]

Can't measure DC with a scope clamp on. (usually)

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[angryman]

this is a handheld meter with a scope in it...just 2 leads like a regular DMM...

http://us.st11.yimg.com/us.st.yimg.com/I/webtronics_1914_4938623

 

[itsmoked]

Just so you know 'too big' caps blow the rectifiers at start up and blow the regulator on an ouput short circuit.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[BrianR]

Adding C causes larger current pulses from the transformer and rectifier which increases heat in the components approx as i^2. nbucksa has given you the accurate equations but I use a rule of thumb to size the C. When 2*pi*f*C*Rl <= 10 then transformer and rectifier heating are tolerable and there is good ripple level. (f=2x mains frequency for a full wave rectifier and Rl is the V/I drawn by the load).

When you make C so large that the figure is much greater than 10 then you will gain very little in terms of ripple reduction but pick up a lot of heat.

This is a crude method but yields very workable results.

 

[angryman]

very helpful info...the only cap I could find was teh 1KuF, and considered throwing another one in parrallel, but ran out of time....

While I've got your ears...sorry...i coudl use a good link to learn more about swtiching 110vac with relays, and using 110vac relay coils...would a cap(s) placed in series with relay contacts (passing ac)help reduce the bounce or spike voltage?

AC Electronics is like Aikido...The more I do, the less I seem to know

 

[BrianR]

Choosing a relay is straight forward provided you understand what the load reactance is. Study the manufacturers data which will typically show that relays rated at 10A for resistive will only be rated at 3A for inductive loads and 3-4A for capacitive. Loads made up of incandescant globes have very high resistive inrush current.

To remove switching transients the most common approach is to use a snubber or MOV or Transzorb etc across the load. However another way is to choose a solid state relay that has zero current switching for inductive loads or zero voltage switching for capacitive loads.

 

[OperaHouse]

If you have an extra transformer, condsider placing the low voltage secondary betwen the rectifier and the filter cap (no primary connection). This will operate as a choke input and reduce the peaks of the output voltage. A cap will increase the measured voltage, but it can still be a very bumpy ride for whatever is being powered. Some devices may not like it. Peak voltage can be 1.4 times the the RMS AC (36V) which could cause a problem. A choke input resists this peak and tries to supply current from the stores magnetic field as the voltage drops. Suitable inductors can also be found in old sodium and mercury lamps.

 

[BrianR]

Operahouse, an interesting idea. The peak voltages are generally much larger than 1.4xRMS due to L di/dt and resonance effects.

 

[angryman]

If I am using a 230vac relay and want to pass the 230vac through the coils as well (same supply) could there be trouble?
Basically, I've got a circuit where I am latching one of the (2) coils with a pushbutton, and run the feedback wire through contacts on the other relay, which is latched with a seperate button (never at the same time).
I don't think solidstate relays are a possiblity...the 'loads' will be almost all inductive....

 

[BrianR]

This is an everyday task and if you are not sure what relay/contactor to use then seek advice from your electrical supplier. If transients are a problem fit a snubber or MOV across the load.