Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives :: Books
Support This Site: Contributors :: Advertise


It is currently Dec Fri 19, 2014 9:01 am


All times are UTC [ DST ]



Post New Topic Post Reply  [ 21 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Question about replacing selenium rectifier
PostPosted: Feb Sat 11, 2012 11:08 pm 
Member

Joined: Jan Sat 14, 2012 11:48 pm
Posts: 49
My Zenith Model L721 has a selenium rectifier with a current rating of .074A.

I have been unable to find a rectifier with that rating. Based on an article I found, I'm planning to replace it with a 1.0A Vishay diode #1N4007, rated at 1000 volts. This will be wired in series with a 200 Ohm, 5W resistor.

I'm a rank radio amateur. Any problems with that?

Thanks. Lucky


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sat 11, 2012 11:28 pm 
Member
User avatar

Joined: Apr Fri 03, 2009 10:32 am
Posts: 1868
Location: New Rochelle, NY
Try this article first:

http://www.w3hwj.com/index_files/RBSelenium2.pdf


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sun 12, 2012 2:36 am 
Member

Joined: Jun Fri 19, 2009 6:34 pm
Posts: 4376
Location: Long Island
True, you will not find a silicon power diode rated for 74 milliamps. They just don't exist. But a 1N4007 will make a fine substitute.

As for the 200-ohm resistor in series, you may want to re-think that a little. There may already be a series resistor in the radio; they were commonly provided to protect selenium rectifiers from turn-on surge. Adding more resistance may lower the B+ voltage too much and cause the radio to perform poorly. It may be that you don't need another resistor, or its value could be something other than 200 ohms. The easy way to tell is get the radio playing at loud volume and check the B+ voltages. If they match those on the schematic within a few volts, everything is fine. If not, the value of the series resistor can be modified as needed to bring the voltages into line.

The other thing--and this is a serious omission in the article cited above--is that when you install a silicon rectifier in a radio that didn't originally have one to begin with, consider what will happen if the diode shorts out. This is because shorting out is the way a silicon diode normally fails, if it ever does. If the result would be full line voltage across the filter cap, it will promptly go "bang" and emit lots of stinky white smoke. In this case. your 200-ohm wirewound series resistor might take much of the current instead and turn cherry red, perhaps damaging other components and wiring around it.

For this reason, I consider fusing mandatory whenever a silicon diode is used. For example, if the input filter is 30 uF, it would have a reactance of 88 Ohms at 60 Hz. Should the diode short (and there be no resistor to limit the current), 1.3 amps would flow through it until it blows up. Therefore a fuse of one amp or less would suffice to protect against diode failure.
Another approach would be to make the series resistor act as a fuse by picking its resistance and wattage values so that it carries the normal load within its ratings, but it would be severely overloaded and quickly burned out if the diode shorts. In this case, a modern flameproof film resistor would be used.

_________________
"Hell, there are no rules here--we're trying to accomplish something!"

Thomas A. Edison


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sun 12, 2012 2:51 am 
Member

Joined: Aug Wed 24, 2011 4:35 am
Posts: 2439
Location: Sunnyvale CA
Chris108 wrote:
True, you will not find a silicon power diode rated for 74 milliamps. They just don't exist. But a 1N4007 will make a fine substitute.

As for the 200-ohm resistor in series, you may want to re-think that a little. There may already be a series resistor in the radio; they were commonly provided to protect selenium rectifiers from turn-on surge. Adding more resistance may lower the B+ voltage too much and cause the radio to perform poorly. It may be that you don't need another resistor, or its value could be something other than 200 ohms. The easy way to tell is get the radio playing at loud volume and check the B+ voltages. If they match those on the schematic within a few volts, everything is fine. If not, the value of the series resistor can be modified as needed to bring the voltages into line.


I would be almost certain that it will need some additional series resistance. The selenium rectifier is dropping something like 20 (or more) volts and the diode will drop around a volt.

Another nasty little issue is that with the selenium rectifier in there, when you first turn it on, the voltage across the filter caps will be something like 130 volts. This is when it is drawing essentially nothing out of the B+. So a 160V filter cap is within its rating. With the silicon diode that goes right up to ~156 volts, which seems a little close for comfort. Once it warms up and starts playing it drops but for the first 15-20 seconds its pretty iffy. In this case it doesn't matter what the series resistance is, there's no current to speak of. Once it warms up the resistor will knock that down to around 100-110V (which is what it is supposed to be, more or less). I have started using either 250V or 450 V filter caps in these sort of radios just to be safe.

Brett


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sun 12, 2012 2:53 am 
Member

Joined: May Tue 30, 2006 4:46 pm
Posts: 6421
Location: Santa Rosa, CA
Chris,

good point about using a fuse or fuse resistor.

Have you ever seen a selenium rectifier while in the process of failing? Lots of nice noxious smoke and an unforgettable smell of burning garlic. I used to work in a TV/Radio service shop in the late 60s and saw a few failures. The result is a lot worse than a silicon diode shorting. And that catastrophe is not prevented by the series resistor... it merely prolongs the cooking.

If I ever do a re-write of the article, I will make the point about preparing for failure. That said, the chances of a 1N4007 failing are less than the chances of an electrolytic cap faililng... which could also do some serious damage in an un-fused circuit.

Check inside a modern (cheap) CFL lamp. You will find a quarter watt resistor that is supposed to fail when the electrolytic cap fails. The electrolytic caps have a very high failure rate in CFLs.

Rich


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sun 12, 2012 2:58 am 
Member

Joined: May Tue 30, 2006 4:46 pm
Posts: 6421
Location: Santa Rosa, CA
Quote:
With the silicon diode that goes right up to ~156 volts, which seems a little close for comfort.


For 120V rms line, the peak voltage is 1.414 x 120 = 169 Volts. Maybe a volt for diode drop. Depending on your series resistor, your 160V cap is already in trouble.

For a reliable conversion, I use 200V caps.

Rich

PS: Selenium cells typically drop about 1V per plate. Most designs were 5 or 6 plates for 110V line use.


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sun 12, 2012 3:10 am 
Member

Joined: Aug Wed 24, 2011 4:35 am
Posts: 2439
Location: Sunnyvale CA
Rich, W3HWJ wrote:
Quote:
With the silicon diode that goes right up to ~156 volts, which seems a little close for comfort.


For 120V rms line, the peak voltage is 1.414 x 120 = 169 Volts. Maybe a volt for diode drop. Depending on your series resistor, your 160V cap is already in trouble.


110*sqrt(2)=155.5... My line voltage is around 110-113 almost all the time, except for the middle of the night. Just now (since I have one on the workbench with this arrangement ready to measure) I got a "cold" voltage across the cap of 162V. Fortunately it's a 250V cap. Oh, and it's 330 mfd instead of 40 in the original, but that's a different thread...

The old selenium rectifier was down to around 120 in the same conditions, which is why the radio didn't work so well.

But, yes, my point exactly, you need more than a 160V cap for reliable operation - even though it's well less than that with the selenium rectifier and when it is operating most of the time. I keep seeing people talking about slapping in diodes and a series resistance without mentioning this issue, so I thought I would. I do agree with replacing the selenium but it's worth mentioning that you might have to do more than just get the warmed-up B+ voltage correct.

And of course, it needs a fuse in any case.

Brett


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sun 12, 2012 4:31 am 
Member

Joined: May Tue 30, 2006 4:46 pm
Posts: 6421
Location: Santa Rosa, CA
My line voltage, according to my new PG&E "Smart Meter," is 122 V.

Haven't seen 110 around here, even during A/C season!

Rich


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sun 12, 2012 5:36 am 
Member

Joined: Jun Fri 19, 2009 6:34 pm
Posts: 4376
Location: Long Island
A couple of thoughts on this.

First, it is only possible for a power supply filter capacitor to charge up to the maximum peak voltage in the abscence of a load. Nearly all modern electrolytics are designed to tolerate brief surges of 10% over their nominal voltages (check the manufacturer's spec sheets for the particular caps you are using), so a 160-volt cap is not "in trouble" if the peak voltage hits 160 volts for a few seconds during warm-up. If it is still a concern, a bleeder resistor that draws a couple of mA at 160 volts, eg. 80-k Ohms at 1 Watt, will reduce the peak voltage across filter capacitors of ordinary radio size.

Under load, the voltage across a filter capacitor falls to a level determined by the size of the capacitor and the current taken by the load. This means that the output voltage can be controlled within a limited range, by increasing or decreasing the size of the filter capacitors. If you only need to reduce the output voltage a handful of volts, a smaller input filter capacitor is often a better choice than adding more resistance in series with the diode.

_________________
"Hell, there are no rules here--we're trying to accomplish something!"

Thomas A. Edison


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sun 12, 2012 7:49 am 
Member

Joined: Aug Wed 24, 2011 4:35 am
Posts: 2439
Location: Sunnyvale CA
Chris108 wrote:
A couple of thoughts on this.

First, it is only possible for a power supply filter capacitor to charge up to the maximum peak voltage in the abscence of a load. Nearly all modern electrolytics are designed to tolerate brief surges of 10% over their nominal voltages (check the manufacturer's spec sheets for the particular caps you are using), so a 160-volt cap is not "in trouble" if the peak voltage hits 160 volts for a few seconds during warm-up. If it is still a concern, a bleeder resistor that draws a couple of mA at 160 volts, eg. 80-k Ohms at 1 Watt, will reduce the peak voltage across filter capacitors of ordinary radio size.

Under load, the voltage across a filter capacitor falls to a level determined by the size of the capacitor and the current taken by the load. This means that the output voltage can be controlled within a limited range, by increasing or decreasing the size of the filter capacitors. If you only need to reduce the output voltage a handful of volts, a smaller input filter capacitor is often a better choice than adding more resistance in series with the diode.


Thanks, I understand all of this. I merely note that the rating is exceeded - and it's more than a few seconds, it's 15-20 second per cold start, and, as you note, upwards of 170 volts worst-case. I have a seen a bunch of them that have survived this so it's not instant death but *I still think it is wise to consider the problem* and to, since we are partially redesigning the power supply anyway, *rate the parts appropriately*.

Brett


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sun 12, 2012 1:19 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 4113
Location: Gainesville, Florida
good information. I saved the article in Hints and Kinks

_________________
CAUTION:Im No Expert
DISCLAIMER:Not responsible For Knowledge or Lack thereof
WARNING Will Robinson


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sun 12, 2012 6:22 pm 
Member

Joined: May Tue 30, 2006 4:46 pm
Posts: 6421
Location: Santa Rosa, CA
Quote:
Nearly all modern electrolytics are designed to tolerate brief surges of 10% over their nominal voltages (check the manufacturer's spec sheets for the particular caps you are using), so a 160-volt cap is not "in trouble" if the peak voltage hits 160 volts for a few seconds during warm-up.


My background is related to electronics in high production manufacturing. My conservative mentality says that I would never use a part that would EVER see its maximum ratings. We were taught to "de-rate" components so that stress is minimized for reliability sake. The higher the voltage across a capacitor the shorter will be its life. Likewise with temperature. Voltage and temperature are "drivers" of insulation failure, ion migration, etc.

Would you use a 1N4007 in a power supply where 950V is repeatedly applied, or even occasionally applied? To replace a 5Y3 for example, I use two 1N4007s in series... just for added safety margin. That's just me... conservative.

Rich


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sun 12, 2012 8:55 pm 
Member

Joined: Jun Fri 19, 2009 6:34 pm
Posts: 4376
Location: Long Island
Hi Rich,

I think it is well to bear in mind that many of the radio power supply filter capacitors discussed in this thread were originally rated for 150 volts, and in some cases even less. Now if you take into account that the silicon diode will drop one volt instead of the five volts dropped by the selenium when it was new, and the fact that the nominal line voltage in the US is 125 volts instead of 120 as it was when the radio was made, you're right back to where you started from.

Using higher voltage capacitors is certainly not going to hurt anything as long as they physically fit, but over time they will simply turn into 160-volt caps anyway and nothing will have been gained by their usage. However, for the reasons given above, I would not advocate using 150-volt or lower replacement caps, even if they were still available.

_________________
"Hell, there are no rules here--we're trying to accomplish something!"

Thomas A. Edison


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Sun 12, 2012 10:44 pm 
Member

Joined: May Tue 30, 2006 4:46 pm
Posts: 6421
Location: Santa Rosa, CA
Quote:
many of the radio power supply filter capacitors discussed in this thread were originally rated for 150 volts


AA5 manufacturers of the 40s and 50s used the cheapest caps available. Old-time cap manufacturers probably over-designed their caps because they didn't have computerized design tools and couldn't completely control the materials they used. It worked back then... with 110V rms line voltage.

You won't find a 160V electrolytic in the cheapest CFL lamp today. They are all 200V or more. I have autopsied quite a few.

Anyone who would leave a 40 year old 150V electrolytic in a "repaired" radio is asking for trouble.

No argument with your original premise... much better to anticipate potential failure and minimize damage.

When I was in the business, our diodes never exploded. We had "spontaneous dis-assembly, accompanied by flame."

Rich

PS: as far as radios go, the only ones that used selenium rectifiers were the AM/FM series string types in the 50s and early 60s. Most other selenium applications had a power transformer that limited available fault current.


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Mon 13, 2012 12:33 am 
Member

Joined: Jun Fri 19, 2009 6:34 pm
Posts: 4376
Location: Long Island
Hi Rich,

This is turning out to be a good thread and I think it is worth carrying it a little further.

Quote:
[...] You won't find a 160V electrolytic in the cheapest CFL lamp today. They are all 200V or more. I have autopsied quite a few.


The last time I checked an electronic CFL that had been on for a while, it was too hot to touch. I wouldn't be the least bit surprised if the internal temperature was over 60 degrees celcius. The voltage ratings of electrolytics (and tantalums for that matter) are determined by the maximum leakage current that can be tolerated before the capacitor literally cooks itself to death. If you've got 140 or 150 volts of rectified DC and the temperature is over 60 degrees C., there's no way a 160-volt cap would hold up. But that's not be the kind of environment expected under the chassis of a radio like the OP's Zenith AM-FM.

Quote:
Anyone who would leave a 40 year old 150V electrolytic in a "repaired" radio is asking for trouble.


No argument there, but consider--hard as it is to do--that 40 years ago is the early 1970s. I think the real "bottom of the barrel" was from the late-1950s to the late-1960s, when US radio manufacturers were trying to hang on in the face of increasing competition from Asia. By the '70s, the US companies had for the most part thrown in the towel on radios, and the tube era was over. Many of the high quality solid state radios that came along around that time still work very nicely with all their original electrolytics still in place.

I recently came across a very well used, mid-1960s GE clock radio that is still playing well on its original orange "dynamite stick" filter capacitor. I considered that so unusual, I put it aside without changing it. To me, that is a curiosity, not a radio I ever intend to use. Should I decide differently in the future, a new pair of caps will of course be installed.

Quote:
When I was in the business, our diodes never exploded. We had "spontaneous dis-assembly, accompanied by flame."

Rich

PS: as far as radios go, the only ones that used selenium rectifiers were the AM/FM series string types in the 50s and early 60s. Most other selenium applications had a power transformer that limited available fault current.


Seleniums turned up in a great many "3-way" battery portables, where they provided B+ as well as filament power when the radio was on AC; and also in "tin can" and "briefcase" TV sets. The TVs were particularly notorious for having transformerless power supplies with voltage doubler circuits. Talk about cheap design!

_________________
"Hell, there are no rules here--we're trying to accomplish something!"

Thomas A. Edison


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Mon 13, 2012 2:20 am 
Member

Joined: May Tue 30, 2006 4:46 pm
Posts: 6421
Location: Santa Rosa, CA
Whoops!

Never say "only." I forgot about portables that used selenium in the "battery eliminator."

Worked on a few zapped selenium doublers in cheap TVs. UL wasn't too knowledgeable about electronics in those days!

Rich


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Mon 13, 2012 4:13 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 12006
Location: Warner Robins, GA
If the selenium still works properly no need to change it IMHO.

Also there are some devices such as a Gray Audograph that run the tube heaters and various coils on the B+ voltage where the selenium rectifier is the only thing that will work properly unless one builds a voltage regulator circuit for it.


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Mon 13, 2012 5:46 pm 
Member

Joined: Nov Sat 27, 2010 6:15 pm
Posts: 4942
Pennies. I keep 450WVDC electrolytics and 630 VDC mylars in common values, and use for everything. Seleniums always get replaced and tamed with dropping resistors. Carbon resistors also get replaced. I reckon about $7.00 or so for typical AA5 a little more if the micas are toast.

Don't know what the issue is.


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Mon 13, 2012 8:28 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 12006
Location: Warner Robins, GA
I've changed a few seleniums before, but if they still work properly and B+ is where it should be then why replace them? Sure they may fail sometime in the future, but that may be years from now.

Only thing about replacing a selenium with a diode and resistor is you now have more resistance in the B+ circuit which will cause more of a variance in B+ voltage as B+ current varies. The voltage variance may not be that much on a simple AA5, but on something that uses coils operated off B+ the voltage variance will be too great for the device to properly operate. Tried that myself on the Gray Audograph I mentioned in my previous post and decided to just leave the selenium alone since it is working perfectly.


Top
 Profile  
 
 Post subject: Re: Question about replacing selenium rectifier
PostPosted: Feb Mon 13, 2012 9:05 pm 
Member

Joined: May Tue 30, 2006 4:46 pm
Posts: 6421
Location: Santa Rosa, CA
Quote:
I've changed a few seleniums before, but if they still work properly and B+ is where it should be then why replace them? Sure they may fail sometime in the future, but that may be years from now.


Do you also leave 50 yr. old electrolytics un-touched?

Quote:
Only thing about replacing a selenium with a diode and resistor is you now have more resistance in the B+ circuit


Actually, you aren't changing the resistance very much. A silicon 1N4007 has an equivalent resistance of about 1 ohm. (1 Volt drop @ 1 amp). A new selenium, say 6 plates rated at 100 mA, has about 60 ohms equivalent resistance (6 volts @ 100 mA.)

Sorry, I am not familiar with the Audograph, so I am sure there are special circumstances.

Rich


Top
 Profile  
 
Post New Topic Post Reply  [ 21 posts ]  Moderators: Marcc, Norm Leal Go to page 1, 2  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: AndrewOl, glasdave, Google [Bot] and 8 guests



Search for:
Jump to:  
















Privacy Policy :: Powered by phpBB