
Page 1 of 1

[ 10 posts ] 

Author 
Message 
hunter2115

Post subject: Resistor wattage for selenium rectifier replacement Posted: May Tue 15, 2012 3:23 pm 

Member 

Joined: Dec Thu 29, 2011 8:45 pm Posts: 1660 Location: York, PA

I know that theres been much discussion on replacing selenium rectifiers with a silicon diode and resistor already, but I cant seem to nail down what wattage rating should be used for the resistor. Some say 2 watts, others say 5 watts or even 10 watts. Is there a general guideline as to what wattage is best considering that space is often limited within a chassis? Bill W
_________________ Bill W


Top 


codefox

Post subject: Re: Resistor wattage for selenium rectifier replacement Posted: May Tue 15, 2012 3:41 pm 

Joined: Nov Sat 27, 2010 6:15 pm Posts: 5430

You're not going to like this, but "It depends."
Determine what the maximum B+ load current will be and use Mr. Ohm to calculate wattage. W=IE. after E=IR where you have determined E and R.
It is typical in old radios to use a marginal wattage resistor at the front of the voltage divider as a cheap fuse that would prevent fires and lawsuits. True. Without them, many old sets would have turned into crispy critters long ago.


Top 


Eickerman

Post subject: Re: Resistor wattage for selenium rectifier replacement Posted: May Tue 15, 2012 3:43 pm 

Member 

Joined: Jan Fri 06, 2012 8:47 pm Posts: 6060

hunter2115 wrote: I know that theres been much discussion on replacing selenium rectifiers with a silicon diode and resistor already, but I cant seem to nail down what wattage rating should be used for the resistor. Some say 2 watts, others say 5 watts or even 10 watts. Generally either 2 or 5 watts should usually be fine, but below the reason for the range of answers is explained below. Here is the deal. It depends upon how much current is being drawn by your radio and how much resistance you are using. It comes from the power calculation which is P = I * I * R where P is power in watts, I is current in amps, and R is resistance in ohms. Whatever you calculate as the power being dissipated in the resistor you at least double it for a resistor "rating". So, lets say your radio pulls 100 mA from the B+ line and you find that a 200 ohm resistor gives you the proper voltage to the rest of the radio. Then P = 0.1 * 0.1 * 200 = 2 watts . So it would be best to use a resistor rated at least 4 watts, so 5 watts is probably a conveniently available part. Lets say that instead your radio only draws 75 mA and you discover 150 ohms gives you the best results for voltage. Then p = .075 * .075 *150 = 0.84 watts or almost a watt. In that case 2 watts would be the recommended standard resistor wattage. Lets say your radio draws 150 mA and you discover you need a 220 ohm resistor for some reason. Then P = .15 * .15 * 220 = 4.95 watts or almost 5 watts. In this case a 10 watt resistor would be recomended. So there you have it. Everything from 2 to 10 watts is the right answer depending upon the question. Curtis Eickerman
_________________ http://curtiseickerman.weebly.com


Top 


Rich, W3HWJ

Post subject: Re: Resistor wattage for selenium rectifier replacement Posted: May Tue 15, 2012 4:15 pm 

Joined: May Tue 30, 2006 4:46 pm Posts: 9067 Location: Santa Rosa, CA

The above advice is good, especially the point about using a resistor with a wattage twice what you have calculated. If you look at a resistor data sheet, the maximum watts rating is usually given for a body temperature well above 100 C and many are spec'd for over 200 C. I wouldn't want anything that hot in my radio or test equipment. A larger resistor has more mass and surface area, so will dissipate heat more efficiently, and run cooler. The goal in selecting a resistor is to maintain the B+ voltage as close as possible to the original spec value. Once you know the resistance, the second step is to calculate the power dissipation, double it, and choose the nearest (higher) wattage resistor. The calculation is not usually 100% accurate, because the rectifier only conducts current for a portion of each AC cycle and the peak current can be 5 or 10X the average value, but it's a good start. You can always adjust the resistor value (select a new resistor) once you measure the actual B+ voltage. http://www.w3hwj.com/index_files/RBSelenium2.pdfRich


Top 


codefox

Post subject: Re: Resistor wattage for selenium rectifier replacement Posted: May Tue 15, 2012 4:27 pm 

Joined: Nov Sat 27, 2010 6:15 pm Posts: 5430

Agree. Add a fuse if you do this.


Top 


Ancient_Hacker

Post subject: Re: Resistor wattage for selenium rectifier replacement Posted: May Tue 15, 2012 5:42 pm 

Joined: May Sat 22, 2010 4:42 pm Posts: 2987

Hmm, well, it's REALLY hard to come up with an exact resistor wattage, as the rectifier circuit draws current in short but high peaks. You really can't use the average current to come up with a reasonable number for the wattage.
You see the wattage goes up as the SQUARE of the current, so if you measure an average current of 1 amp thru a 1 ohm resistor, you'd think the wattage would be 1 squared times 1, or 1 watt.
But in a real rectifier circuit the real current only exists for like 1/5 of a cycle, and in this example, it would be at 5 amps, so the calculation is 5 squared times 1, or 25, but only on for 1/5 of the time, so divide by 5: 5 watts. Quite a different answer.
In your typical old tube radio halfwave rectifier, the current is more like my example than the average current example. So it's best to assume like 4 o5 5 times the current for 1/4 or 1/5 of a cycle. You get much more realistic wattage values that way.


Top 


Tube Radio

Post subject: Re: Resistor wattage for selenium rectifier replacement Posted: May Tue 15, 2012 6:00 pm 

Member 

Joined: Jan Thu 01, 1970 1:00 am Posts: 17483 Location: Warner Robins, GA

If this is not for a vacuum tube operated TV and the selenium is still producing the correct voltage then IMHO there is no need to replace it.


Top 


sofaslug

Post subject: Re: Resistor wattage for selenium rectifier replacement Posted: May Tue 15, 2012 6:05 pm 

Moderator 

Joined: Jan Thu 01, 1970 1:00 am Posts: 17455 Location: Berkeley, CA 94709

Ancient_Hacker wrote: Hmm, well, it's REALLY hard to come up with an exact resistor wattage, as the rectifier circuit draws current in short but high peaks. You really can't use the average current to come up with a reasonable number for the wattage.
You see the wattage goes up as the SQUARE of the current, so if you measure an average current of 1 amp thru a 1 ohm resistor, you'd think the wattage would be 1 squared times 1, or 1 watt.
But in a real rectifier circuit the real current only exists for like 1/5 of a cycle, and in this example, it would be at 5 amps, so the calculation is 5 squared times 1, or 25, but only on for 1/5 of the time, so divide by 5: 5 watts. Quite a different answer.
In your typical old tube radio halfwave rectifier, the current is more like my example than the average current example. So it's best to assume like 4 o5 5 times the current for 1/4 or 1/5 of a cycle. You get much more realistic wattage values that way. What Ancient Hacker says is what I've found in my experience when determining dropping resistances in rectifier circuits. Simply using a 2x factor will result in the resistor turning into a miniature space heater. 10x is more realistic (2x times the 5x noted in his example). Even then you'll likely get a pretty toasty resistor, but at least one that couldn't be used to fry an egg. Bob


Top 


Eickerman

Post subject: Re: Resistor wattage for selenium rectifier replacement Posted: May Tue 15, 2012 6:33 pm 

Member 

Joined: Jan Fri 06, 2012 8:47 pm Posts: 6060

You can always use more than 2X of the calculated dissipation for any resistor. This is just the minimum that should provide reliable service over a very long time (measured in years). Of course they WILL get hot.
The bigger the component, the less the surface area temperature for the same amount of power being dissipated.
100 watts dissipated at the small tip of a soldering iron will melt solder. 100 watts dissipated over a surface the size of a lightbulb will not. Either way it is 100 watts. It is the same with resistors. Physically bigger resistors dissipating the same amount of energy will run cooler. They will still heat the underside of a chassis the same because 5 watts is 5 watts.
Curtis Eickerman
_________________ http://curtiseickerman.weebly.com


Top 


Rich, W3HWJ

Post subject: Re: Resistor wattage for selenium rectifier replacement Posted: May Tue 15, 2012 7:37 pm 

Joined: May Tue 30, 2006 4:46 pm Posts: 9067 Location: Santa Rosa, CA

I think the 2x rule on wattage is pretty reasonable.
Another way to look at it: the dropping resistor is adding the voltage drop that formerly took place within the selenium rectifier. A typical selenium might have 1 inch plates and dissipate a couple of watts at most. Properly sized, seleniums don't get very hot. If you are replacing a large selenium, say 2 inch or greater plates, you may need a large watt resistor.
The dropping resistor shouldn't be adding any heat to the system... just moving it from the rectifier to a resistor.
Rich


Top 



