Tech: early Bally Rectifier boards and hot resistors
GPE
OK -- the 25 and 600 ohm resistors on these rectifier boards naturally run
hot - but do they really need to? I'm trying to reduce the heat generated
by these two.
Under normal operations:
The 25 ohm resistor is constantly dissipating 1.62 watts (constant voltage
of 6.36)
The 600 ohm resistor is constantly dissipating 3.48 watts (constant voltage
of 45.7)
When using 25 & 600 ohms -- I got 55C/131F and 84C/183F, respectively.
I tried using 33 and 750 ohm resistors and don't see any detrimental effects
of the rectifier board.
Now, the 33 ohm resistor is constantly dissipating 1.23 watts (constant
voltage of 6.36)
and the 750 ohm resistor is constantly dissipating 2.78 watts (constant
voltage of 45.7)
These power ratings still are enough to generate a lot of heat.
When using 33 and 750 ohms -- I got 46C/114F and 69C/156F, respectively.
However, my temperature probes freaked out somewhere during testing.... so
I'm not sure on the accuracy of *any* of these measurements. Halfway thru
testing I discovered I can get whatever temperature I want by merely
touching the cable.... stupid POC.
What I want to do is get the temperature down to half of the original temps.
Easiest way that I can see is to double the resistance -- from 25 & 600 to
47 & 1.3K (based on resistor values I have).
I can't try this until I get a new temperature probe.
But -- has anybody out there tried replacing these resistors with higher
values?
If so -- what were your results? Good, bad, indifterent?
TIA,
Ed
kenn...@yahoo.com
First off, loose the mechanical temperature probe. I got a nice Fluke
IR meter for $100 and it goes to 600 F with a max reading hold. It
just get it close to the surface I want to measure and kind of wave it
around. Since the resistor will be the hottest thing you point it at,
the max reading will be the true heat of the surface of the ceramic.
Very repeatable too.
I believe R1 is how they got the ac ripple induced with a minimum
guaranteed load onto the 43 volt solenoid supply. I would suspect it
acts as a voltage divider, so you could use a Thevinen equvilent
resistance if you know the voltage across and current through the
resistor (I'm thinking one end of the thevinin model is ground, the
other is the output of the resistor/bridge node). You do know both of
those values, so you can predict the system changes on the Thevinin
circuit by modeling the new resistor value. I would suspect as long as
the zero cross circuit still works, any lower resistor value should be
OK.
I think the 25 ohm resistor is the same thing. They wanted to make
sure that at zero crossing the output of the bridge goes to 0 volts.
Figure an average current of 3 amps(20 lamps), the approximate load
resistance is 2.5 ohms. A factor of ten seems a good rule of thumb, so
the 25 ohm resistor make sense. Doubling the value of R2 makes it a
factor of 20. Probably not enough of a difference given the circuit to
be a problem. With fewer lamps on, a higher resistance is probably
less of a factor.
In summary, I don't see how doubling the values would hurt anything,
and if it did, the problems should not be catastrophic and reversable
with a change to lower values.
K2
GPE
<kenn...@yahoo.com> wrote in message
news:1144860571.5...@z34g2000cwc.googlegroups.com...
> Ed,
>
> First off, loose the mechanical temperature probe. I got a nice Fluke
> IR meter for $100 and it goes to 600 F with a max reading hold. It
> just get it close to the surface I want to measure and kind of wave it
> around. Since the resistor will be the hottest thing you point it at,
> the max reading will be the true heat of the surface of the ceramic.
> Very repeatable too.
One of them DOH! momemts. I ditched tie temp probe already -- just ordered
a Fluke 62 IR thermometer. Will redo testing when it arrives. Measures
upto 900F+!! Flame on!!!
>
> I believe R1 is how they got the ac ripple induced with a minimum
> guaranteed load onto the 43 volt solenoid supply. I would suspect it
> acts as a voltage divider, so you could use a Thevinen equvilent
> resistance if you know the voltage across and current through the
> resistor (I'm thinking one end of the thevinin model is ground, the
> other is the output of the resistor/bridge node). You do know both of
> those values, so you can predict the system changes on the Thevinin
> circuit by modeling the new resistor value. I would suspect as long as
> the zero cross circuit still works, any lower resistor value should be
> OK.
Can't be a voltage divider. Only one resistor and it's straight across the
load. But it does make it easy to calculate current draw and power
dissipation. Would be able to calculate the temperature as well IF any of
the resistor manufacturers would give the device to ambient thermal
resistance.
Since there is no capacitive element to this and the resistor draw is
minimal compared to transformer's capability -- this resistor has minimal
affect on output voltage. When varying resistance -- the voltage was
varying by only a couple hundredths of a volt. These variations could
easily have been caused by meter lead connections.
>
> I think the 25 ohm resistor is the same thing. They wanted to make
> sure that at zero crossing the output of the bridge goes to 0 volts.
> Figure an average current of 3 amps(20 lamps), the approximate load
> resistance is 2.5 ohms. A factor of ten seems a good rule of thumb, so
> the 25 ohm resistor make sense. Doubling the value of R2 makes it a
> factor of 20. Probably not enough of a difference given the circuit to
> be a problem. With fewer lamps on, a higher resistance is probably
> less of a factor.
My guess is that it has something to do with the zero crossing as well.
(Being FWB rectified, there really are no zero crossings -- always either a
brief zero or a positive voltage)
However, I don't see how the number of lamps has anything to do with it. If
the lamps (or solenoid) are on - you have a parallel load. If it was a
serial load - I could see this making a difference.
Where I can see a problem would be if -no- loads are turned on. If you
didn't have a load resistor and no loads were turned on -- the slightest bit
of stray capacitance (i.e. wiring) will serve as a minute filter cap. This
could cause a lack of zero crossings. The load resistors would overcome any
possibility of stray capacitance and guarantees zero crossing detections.
>
> In summary, I don't see how doubling the values would hurt anything,
> and if it did, the problems should not be catastrophic and reversable
> with a change to lower values.
> K2
I agree. To me, it still seems to be an arbitrary selection on resistors.
My guess is you could actually go 10x or more with the resistors and still
not have a problem. When I get my new temp probe in - I'll try the 2x
resistors.
-- Ed