Maker Pro
Maker Pro

Bridge rectifier question

D

Dan Beck

Jan 1, 1970
0
Hello all,

I have a 35A, 400V bridge rectifier that may be the source of a problem.
When I use the diode test on my DMM, with the power to the bridge off, the
forward voltages measure 0.470 V, which I thought was normal. The AC input
to this bridge is 29 VAC; I am getting 36 VDC output at the DC leads. Is
this normal, does this represent a problem with the bridge, or with the DC
circuitry downstream?

Thank you for reading.
Dan
 
J

John Popelish

Jan 1, 1970
0
Dan said:
Hello all,

I have a 35A, 400V bridge rectifier that may be the source of a problem.
When I use the diode test on my DMM, with the power to the bridge off, the
forward voltages measure 0.470 V, which I thought was normal. The AC input
to this bridge is 29 VAC; I am getting 36 VDC output at the DC leads. Is
this normal, does this represent a problem with the bridge, or with the DC
circuitry downstream?

Thank you for reading.
Dan

An AC voltage that measures 29 volts on a volt meter produces the same
heating effect as a DC voltage of 29 volts if applied across a
resistor. But AC sine wave has two moments per cycle of zero volts
and two other moments per cycle that are the square root of 2 ( or
1.414) times this effective voltage. For 29 volts AC, the peak
voltage is about 41 volts. When you take a DC reading of the output
of the bridge, you measure the average voltage, not the effective
voltage the AC meter measures. The average of a rectified sine wave
that has a peak voltage of 41 volts is 63.7% of the peak or about 26
volts. If you are measuring something much higher than that, there
must be a capacitor after the bridge that is charging up to almost the
peak and discharging very little between peaks.

see:
http://www2.sfu.ca/sonic-studio/handbook/Root_Mean_Square.html
 
B

Bill Vajk

Jan 1, 1970
0
John said:
Dan Beck wrote:
An AC voltage that measures 29 volts on a volt meter produces the same
heating effect as a DC voltage of 29 volts if applied across a
resistor. But AC sine wave has two moments per cycle of zero volts
and two other moments per cycle that are the square root of 2 ( or
1.414) times this effective voltage. For 29 volts AC, the peak
voltage is about 41 volts. When you take a DC reading of the output
of the bridge, you measure the average voltage, not the effective
voltage the AC meter measures. The average of a rectified sine wave
that has a peak voltage of 41 volts is 63.7% of the peak or about 26
volts. If you are measuring something much higher than that, there
must be a capacitor after the bridge that is charging up to almost the
peak and discharging very little between peaks.

Or to put this another way.....

Take the area under the sine wave and redraw it into a square
wave and you get .707 times the peak voltage. That's the
voltage read by your ac voltmeter, the 29 volts.
 
J

John Popelish

Jan 1, 1970
0
Bill said:
Or to put this another way.....

Take the area under the sine wave and redraw it into a square
wave and you get .707 times the peak voltage. That's the
voltage read by your ac voltmeter, the 29 volts.

No. That would be the average value. The effective voltage square
root of the area under the squared sine wave. That is what RMS
means. Root of the mean of the square.
 
B

Bill Vajk

Jan 1, 1970
0
John said:
Bill Vajk wrote:
No. That would be the average value. The effective voltage square
root of the area under the squared sine wave. That is what RMS
means. Root of the mean of the square.

You're generally right, John, but alas I think you're missing
something this time out.

The problem starts with an ac voltmeter reading of 29 volts.

Unless there's something special about the meter it registers
rms voltage for oerdinary sine wave alternating current. I
don't see where you're getting average voltage out of
anything I wrote but I'm certainly willing to listen.
 
J

John Popelish

Jan 1, 1970
0
Bill said:
You're generally right, John, but alas I think you're missing
something this time out.

The problem starts with an ac voltmeter reading of 29 volts.

Unless there's something special about the meter it registers
rms voltage for oerdinary sine wave alternating current. I
don't see where you're getting average voltage out of
anything I wrote but I'm certainly willing to listen.

Perhaps I misunderstood what you meant by "redraw it into a square
wave".
 
F

Fred Stevens

Jan 1, 1970
0
Dan Beck said:
Hello all,

I have a 35A, 400V bridge rectifier that may be the source of a problem.
When I use the diode test on my DMM, with the power to the bridge off, the
forward voltages measure 0.470 V, which I thought was normal. The AC input
to this bridge is 29 VAC; I am getting 36 VDC output at the DC leads. Is
this normal, does this represent a problem with the bridge, or with the DC
circuitry downstream?

Thank you for reading.
Dan

Hi,

Full wave rectification will give an output near the peak voltage of
the AC wave which is 1.4142 times the RMS voltage (29V in your case),
minus the forward drops of the diodes may explain why you get 36V.

Fred.
 
J

John G

Jan 1, 1970
0
Fred Stevens said:
Hi,

Full wave rectification will give an output near the peak voltage of
the AC wave which is 1.4142 times the RMS voltage (29V in your case),
minus the forward drops of the diodes may explain why you get 36V.

Fred.

Fred, If only John P and some others could be as succinct as
you.
All the averages and RMSs etc only serve to confuse someone
asking such a basic question.

Yes a simple rectifier with a capacitor across the output
will give about 1.4 times the transformer AC voltage with no
load and something less as the load increases, depending on
both the capacitor value and the current ratings of the
rectifier and transformer.
 
J

John Fields

Jan 1, 1970
0
Fred, If only John P and some others could be as succinct as
you.
All the averages and RMSs etc only serve to confuse someone
asking such a basic question.

---
Unfortunately, ignoring the details doesn't make him right.
---
Yes a simple rectifier with a capacitor across the output
will give about 1.4 times the transformer AC voltage with no
load and something less as the load increases, depending on
both the capacitor value and the current ratings of the
rectifier and transformer.

---
You may want to go back and read the original post, where no mention was
made about a "capacitor across the output", rendering both your and
Fred's comments useless.

You may also want to go back and read Popelish's first post, where the
possibility of the existence of a filter cap was pointed out, further
nullifying the need for your and Fred's posts.
 
M

Monk

Jan 1, 1970
0
Hello all,

I have a 35A, 400V bridge rectifier that may be the source of a problem.
When I use the diode test on my DMM, with the power to the bridge off, the
forward voltages measure 0.470 V, which I thought was normal. The AC input
to this bridge is 29 VAC; I am getting 36 VDC output at the DC leads. Is
this normal, does this represent a problem with the bridge, or with the DC
circuitry downstream?

Thank you for reading.
Dan

I think those other guys may have already answered your question but
this is good practice for me as im studying power supplies right now
in my devices class. anyways i would say no, there is no problem with
your rectifier.

29 VAC / 0.707 = ~41 Vpeak


41 Vp subtract the voltages from the diode drops and the voltage
variation of the ripple and you should get a voltage DC of about 36.
This really depends of if there is a capacitor there to filter out the
ripple or not but overall it seems like your rectifier is working
properly.

Cheers,
Monk
 
D

Dan Beck

Jan 1, 1970
0
Gentlemen:

Thank you for the prompt, informative responses. I appreciate the
scientific rigor of the answers, even though I had to dust off the calculus
based college physics text from 20 years ago! Sometimes I wonder if I
should have pursued electronics further than that...

The bridge rectifier in question is for a pinball I am working on (my own,
as a hobby). There are a number of feeds off of the DC output; many of them
have filter caps, as Mr. Popelish first speculated. The game has had an
intermittent sound problem I thought related to the output of this bridge; I
believe I need to search a different avenue!

Again, thank you.

Regards,
Dan
 
L

Louis Bybee

Jan 1, 1970
0
John G said:
Fred, If only John P and some others could be as succinct as
you.
All the averages and RMSs etc only serve to confuse someone
asking such a basic question.

Yes a simple rectifier with a capacitor across the output
will give about 1.4 times the transformer AC voltage with no
load and something less as the load increases, depending on
both the capacitor value and the current ratings of the
rectifier and transformer.


I understand, and accept the fact that DC voltage from a rectifier with
capacitance across the unloaded/lightly loaded output can be modeled
mathematically. Is it also possible to model mathematically the DC voltage
from the same output where the load, and capacitance levels are varied?

Thank you.

Louis--
*********************************************
Remove the two fish in address to respond
 
J

John Popelish

Jan 1, 1970
0
Louis said:
I understand, and accept the fact that DC voltage from a rectifier with
capacitance across the unloaded/lightly loaded output can be modeled
mathematically. Is it also possible to model mathematically the DC voltage
from the same output where the load, and capacitance levels are varied?

In principle it is certainly possible to model the rectification and
filtering mathematically. In practice it is pretty messy. Some
complicating factors:
* resistance of transformer windings
* leakage inductance between primary and secondary
* distortion in the line waveform
* non ideal diode drop (diode series resistance)
* non ideal capacitance (ESR and inductance)

So most models are simplified to the point of being only good guesses
at the exact DC and AC ripple components out of any given system. At
least the ideal assumptions put an upper limit on the DC component.

There are quite a few pages on the web on his topic:
http://www.mitedu.freeserve.co.uk/Circuits/Power/unreg.htm
http://my.integritynet.com.au/purdic/power1.html
http://www.toroid.com/custom_transformers/technical_bulletin_1.htm
http://www.onsemi.com/pub/Collateral/HB214-D.PDF
 
L

Louis Bybee

Jan 1, 1970
0
John Popelish said:
In principle it is certainly possible to model the rectification and
filtering mathematically. In practice it is pretty messy. Some
complicating factors:
* resistance of transformer windings
* leakage inductance between primary and secondary
* distortion in the line waveform
* non ideal diode drop (diode series resistance)
* non ideal capacitance (ESR and inductance)

So most models are simplified to the point of being only good guesses
at the exact DC and AC ripple components out of any given system. At
least the ideal assumptions put an upper limit on the DC component.

There are quite a few pages on the web on his topic:
http://www.mitedu.freeserve.co.uk/Circuits/Power/unreg.htm
http://my.integritynet.com.au/purdic/power1.html
http://www.toroid.com/custom_transformers/technical_bulletin_1.htm
http://www.onsemi.com/pub/Collateral/HB214-D.PDF

Hmmmm..... How high is up? :-]

As I expected.

Thank you.

Louis
 
Dan Beck said:
Gentlemen:

Thank you for the prompt, informative responses. I appreciate the
scientific rigor of the answers, even though I had to dust off the calculus
based college physics text from 20 years ago! Sometimes I wonder if I
should have pursued electronics further than that...

The bridge rectifier in question is for a pinball I am working on (my own,
as a hobby). There are a number of feeds off of the DC output; many of them
have filter caps, as Mr. Popelish first speculated. The game has had an
intermittent sound problem I thought related to the output of this bridge; I
believe I need to search a different avenue!

Again, thank you.

Regards,
Dan

if you still come back to this thread
Ok you said sound problem so your machine is solid state for the most
part. But you said it had a transformer so it most be about 20 years
old. You could just buy a new bridge it will only cost you a $1.40-6.
If that dosn't fix it try the caps. You said sound was a problem hmm
might it be just a cap. But .4 volts across the bridge means you have
problems with the bridge so just replace it. It should be with a load
..7-.88 per diode.

what brand is your pinball
is it Stern, Illinois, Capcom, Gottlieb
Williams/Bally, Sega. Or some other brand
 
D

Dan Beck

Jan 1, 1970
0
Hello,

You may regret opening this can of worms, but I would be happy to oblige
your curiosity!

The game is a 1989 Gottlieb System 80B pinball called Big House. I have
performed the following:

Replaced all the electrolytic caps on the sound board power supply board, no
electrolytics on sound board itself.

Repinned the edge connectors carrying the sound lines from CPU board to
solenoid driver board to sound board.

Upgraded the circuit grounds; approximately 40 wires. Gottliebs of this
vintage are notorious for having crummy grounds.

Replaced various TTL chips that Gottlieb gurus suggested I change--no
improvement.

The sound board has a dual CPU system using 2 Rockwell 6502 processors.
Each processor has its own sound rom, amplifier chips, and various TTL
chips; both circuits share a DAC. When certain playfield switches close the
CPU circuit that provides speech cuts out. These playfield switches are
associated with transistors that control 24 (actually 36) volts DC either
through light bulbs or a solenoid. The diode on the solenoid tests ok. I
was thinking the 24 VDC bridge was the troublemaker. This is the bridge
that actually outputs 36 VDC, discussed in my first post. This 36 VDC is
pared down to 12 VDC on the sound board power supply board using a 12 volt
Zener; this 12 VDC then goes to the sound board to power the amplifier
chips. My oscilloscope shows all the address and data lines cut out when
the speech crashes; the other CPU circuit seems immune to the crashes. The
crystal still ticks, and the 5 VDC is good. I have replaced the speech CPU
with a new one--no change.

There are probably other things I have done and not mentioned, but those are
the major ones. Good thing this is a hobby, and not my day job--my family
and I would be starving. Most of the Gottlieb gurus I have spoken with
think it is a flaky sound board; perhaps, but what bothers me is that it
seems an insult external to the sound board is causing the sound board to
crash. Ah well. I have spent way too much of your time; I apologize for
the length. Thank you for humoring me!

Regards,
Dan

P.S. I have a fresh batch of bridge rectifiers. When I test them their
forward voltages (juction drop) are 0.490 volts as well; does this mean they
are bad?
 
The game is a 1989 Gottlieb System 80B pinball called Big House. I have
performed the following:


Hmm model 713
With all the things you have done I guessing you have the manual and
schmatics
http://www.pbresource.com/mansch.html#manaval
Replaced various TTL chips that Gottlieb gurus suggested I change--no
improvement.
TTL are rock steady no ESD issues but and easy fix.

What about the PROMS/roms can go bad in 10 years you may have just
lost some data when a prom goes bad only sections go bad.
http://www.pinballmania.com/memorias.html
http://www.phc.igs.net/~jsj/gottlieb.htm

My oscilloscope shows all the address and data lines cut out when
the speech crashes; the other CPU circuit seems immune to the
crashes.

hm maybe its the eprom
the little guy in my heads says it might be this! If their is one bit
wrong it will crash.
crystal still ticks, and the 5 VDC is good. I have replaced the speech CPU
with a new one--no change.
P.S. I have a fresh batch of bridge rectifiers. When I test them their
forward voltages (juction drop) are 0.490 volts as well; does this mean they
are bad?

No its good there is nothing wrong with you bridge and with a scope
you would now.

I thought that the power supply was supposed to give 30 volts DC to
sound.

Sorry for not being able to help.
 
Top