Apple AirPort Router SMPS, Dummy Load Test - Repair

[SgtRock]

Greetings EEVBees:

--It just occurred to me that some of the projects I have been posting in "General Chat" actually belong in this category. Administration is welcome to move, if deemed necessary. I apologize for being thoughtless.

--Now, back on topic. I was given an Apple AirPort Router 1.8 A 12 V SMPS which was reported to have no output. I cut the 12 V output plug off of the SMPS and soldered it to a working 12 V linear PS, that solved the immediate problem. When I finally got around to checking the output of the SMPS, I found that it was putting out 12 V, and I surmised that it must be a cracked solder joint that was heating up and separating. I managed to cut the darn thing open, which was not easy or pretty, and with a jewelers loupe inspected it. Other than some uncleaned flux, I could find nothing wrong with it. So I sealed it back up using electronic silicone.


 I then hooked it up to power resistors to draw the rated 1.8 A and placed the whole thing in a Pyrex plate. The first picture below shows that the PS is working but is sucking hind teat, and is unable to come up to the full V and A. And, the second picture shows the PS output after two minutes. It has gone completely bonkers and is switching on on off an jumping around like crazy.

--I opened the PS back up and cast a critical eye (see third picture below) to the small capacitor with the black mark on the top. At first I assumed that this was just a check off mark by a Sharpie like the one on the blue rectangular safety capacitor, but now I am not so sure. So I removed all three capacitors marked "Koshin", a brand unknown to me, and have ordered good known SMPS caps from Digi-Key. Also, more for my own edification, that any other reason, since when I open a non-working SMPS I am going to replace all but the largest Caps, I have ordered a Dick Smith Blue ESR Meter Kit.

--Your comments are humbly solicited, and I will report back after attempted repair and testing.

--Would someone please tell me how to move picture into the body of the text? Or can anyone tell me where to look for that information?

"Three weeks in the lab will save you a day in the library every time"
R. Stanley Williams 1951 -

Best Regards
Clear Ether

[vtl]

--Would someone please tell me how to move picture into the body of the text? Or can anyone tell me where to look for that information?

You can post first then edit it to include them once the images have been uploaded. Right click on the picture and copy the link

What does the output waveform look like?

[SgtRock]

Dear Vtl:

--I am afraid I am not having much luck following your picture posting advice. When I right click on the image, nowhere do I see an option for "Copy Link". I see "Save Link As", "Copy Link Location", "Copy Image", "Copy Image Location" and "Save Image As". I have tried "Copy Link Location" and then posting, but all I get is the blue colored link and no picture, as you can see in the original posting.

--I do not have a working scope on hand at the moment, and have already desoldered the caps so I am afraid I cannot tell you what the DC wave form looks like. Thanks again.

"He was born ignorant, and has been losing ground ever since."
Fred Allen 1894 1956

Best Regards
Clear Ether

[vtl]

So right clicking on the image go into properties and copy the target (this is in IE but its similar on chrome and ff):




example link: https://www.eevblog.com/forum/projects-designs-and-technical-stuff/apple-airport-router-smps-dummy-load-test-repair/?action=dlattach;attach=20007;image

[baljemmett]

--I am afraid I am not having much luck following your picture posting advice. When I right click on the image, nowhere do I see an option for "Copy Link". I see "Save Link As", "Copy Link Location", "Copy Image", "Copy Image Location" and "Save Image As". I have tried "Copy Link Location" and then posting, but all I get is the blue colored link and no picture, as you can see in the original posting.

You're almost there (you probably want 'copy image location' for the full-size image, 'copy link location' will give you the thumbnail) -- once you've copied the link you want to paste it inside an img tag like this:


[ img ]https://www.eevblog.com/forum/projects-designs-and-technical-stuff/apple-airport-router-smps-dummy-load-test-repair/?action=dlattach;attach=20054;image[/img]

(There's an 'insert image' button at the left-hand end of the second row of the editor toolbar, or just type img and /img within square brackets and stick the link between them.)

However, if you do this, you might want to go with smaller images, especially in terms of filesize -- if you'd inlined the four images on your original post, for instance, that'd be 2MB of images brought down just viewing the post.  Some people will prefer to click the thumbnails as and when they'd like to view the images, especially for that volume of data; hopefully these people won't mind my above inlining of 150KB for didactic purposes! 

(Edit: I can't type.)

[sonicj]

your syntax should look like this :

Code: [Select]

[img]https://www.eevblog.com/forum/projects-designs-and-technical-stuff/apple-airport-router-smps-dummy-load-test-repair/?action=dlattach;attach=20054;image[/img]

[SgtRock]

Greetings EEVBees:

--Well, I replaced all three of the caps, even though they tested normal for capacitance and ESR.  When I checked the router's actual draw, it was 6.6 Amps, so I soldered up a dummy load to draw 7 Amps and viola (see below picture) it is as stable as a rock and barely warm to the touch. The router is now hooked back up to it and is working fine.

--I would hate to think what would happen if I tried to run it at the Amperage listed on the case. I have read elsewhere that most of the cheaper (I.E One Hung Low) ATX SMPS,s  will explode or at least fail in a very shot time, if run at the wattages printed on the case.

--It is also interesting and a lesson to me, that replacing the Caps, even though they tested good for Capacitance and ESR, made for a good repair.

--Can anyone provide advice on load testing ATX SMPS,s? I want to develop some hard data. I am thinking about using high wattage wire wounds.

--Thanks for all the help with image posting. Now that I understand it, I really effing hate it. I blame SMF software not DJ. Posting images should be drag and drop with auto resizing to forum limits. Images should post in reasonable size, and allow the user to click if he wants the full monty.

“Horse sense is the thing a horse has which keeps it from betting on people."
W. C. Fields (William Claude Dunkenfield) 1880 1946

Best Regards
Clear Ether

[bfritz]

Greetings EEVBees:

--Well, I replaced all three of the caps, even though they tested normal for capacitance and ESR.  When I checked the router's actual draw, it was 6.6 Amps, so I soldered up a dummy load to draw 7 Amps and viola (see below picture) it is as stable as a rock and barely warm to the touch. The router is now hooked back up to it and is working fine.

Please look carefully at your meter in the forth picture.  Do you see the lower case "m" in front of the A on the face of your meter?  Your picture is a bit blurry, bit it looks to me like it is on.  This is the indication that your meter is reading mA.  Note that the 692 is 692mA, which is the same as 0.692A.

There is no way that the little Airport Extreme router will draw 7 amps for a time any longer than a several microsecond surge current.

Note that in the third picture, you are using 3 resistors in series at 10 ohms each.  A total of 30 ohms.  With a 12V output, that would result in a current of:  I = V / R = 12 / 30 = 0.4A, or 400mA.  You can not draw 7A from a 12V output with 30 ohms of load!

Note that picture four shows two of the resistors in series, and then two in parallel that are then put in series with the other two, for a total load of 25 ohms.  This would result in a current of I = V / R = 12 / 25 = 480mA.  The meter appears to be showing 692mA.  Either the resitors are not quite connected as they appear to be, or some of them have been damaged by too much power dissipation, or your meter is not well calibrated.

--I would hate to think what would happen if I tried to run it at the Amperage listed on the case. I have read elsewhere that most of the cheaper (I.E One Hung Low) ATX SMPS,s  will explode or at least fail in a very shot time, if run at the wattages printed on the case.

Overheat or catch on fire, I might believe.  Explode?  I don't think so.

--It is also interesting and a lesson to me, that replacing the Caps, even though they tested good for Capacitance and ESR, made for a good repair.

From what I see in the pictures, I'm not sure your meter is good, or that you aren't doing something strange we don't see.  It appears your meter is saying 692mA, and yet you are discussing 7A.

--Can anyone provide advice on load testing ATX SMPS,s? I want to develop some hard data. I am thinking about using high wattage wire wounds.

--Thanks for all the help with image posting. Now that I understand it, I really effing hate it. I blame SMF software not DJ. Posting images should be drag and drop with auto resizing to forum limits. Images should post in reasonable size, and allow the user to click if he wants the full monty.

If you want to make a very generous donation to Dave for the extra bandwidth, and some of his time to find a change or hack for the forum software, he might jump right on that.  You did donate before complaining didn't you?  It was enough to pay for a day or so of his time, right?

[SgtRock]

Dear Bfritz:

--Indeed, you are correct, I do seem to have slipped a decimal place. And yes I meant to say 0.7 Amps as shown (nearly) in picture 4 (marked "4 Now Working"). I have a severe cold and have been paying less attention than I should have. Mea Culpa. I am only an amateur and, I freely admit, often make stupid mistakes and oversights. As you can easily tell I have not done much of this sort of thing before. I post it so that people who know more than me  can show me the error of my ways (as you have so graciously done) , and to show the specific device for, those who are interested.

--I hope you will not take this wrong, but I believe you are reading  the forth picture (marked "4 Now Working" incorrectly. I think you will find on closer inspection that what is pictured is 2 10 Ohm resistors in parallel, in series with a 2 Ohm resistor and another 10 Ohm resistor, making a total resistance of 5 + 2 + 10, or 17 Ohms nominal. Using Ohms Law I=E/R, I get a computed current of 0.706 Amps nominal. Which is close enough to the meter reading for testing purposes. Once again I must apologize for typing 7 Amps when I meant .7 Amps.

--Not to put too fine a point on it, but your response to my statement about SMF picture posting that "I blame SMF software not DJ.", indicates that my statement may not have been as clear as it should have been. Like I say I have been suffering from a dreadful cold. What I should have said was "I blame SMF software not DJ."

--In any case I want to thank you for correcting my stupid mistakes.

"He was born ignorant, and has been losing ground ever since."
Fred Allen 1894 1956

Best Regards
Clear Ether

[SeanB]

On any SMPS odd outputs are often easy to fix by replacing all electrolytics in it, especially those on the output side, as they often lead a hard life. ESR on those has to be a few dozen milliohms at most, and if they are in any way bulging or swollen, or are discoloured and brown then definitely change them.

[bfritz]

Dear Bfritz:

--Indeed, you are correct, I do seem to have slipped a decimal place. And yes I meant to say 0.7 Amps as shown (nearly) in picture 4 (marked "4 Now Working"). I have a severe cold and have been paying less attention than I should have. Mea Culpa. I am only an amateur and, I freely admit, often make stupid mistakes and oversights. As you can easily tell I have not done much of this sort of thing before. I post it so that people who know more than me  can show me the error of my ways (as you have so graciously done) , and to show the specific device for, those who are interested.

I'm glad you didn't take my comments as being accusational, as they were not meant that way.  We all make errors.  If we learn from them or not is what I believe set the great apart.

--I hope you will not take this wrong, but I believe you are reading  the forth picture (marked "4 Now Working" incorrectly. I think you will find on closer inspection that what is pictured is 2 10 Ohm resistors in parallel, in series with a 2 Ohm resistor and another 10 Ohm resistor, making a total resistance of 5 + 2 + 10, or 17 Ohms nominal. Using Ohms Law I=E/R, I get a computed current of 0.706 Amps nominal. Which is close enough to the meter reading for testing purposes. Once again I must apologize for typing 7 Amps when I meant .7 Amps.

You are correct.  My mistake.  That makes much more sense.  Now we've each had a chance to make a silly mistake!  LOL

--Not to put too fine a point on it, but your response to my statement about SMF picture posting that "I blame SMF software not DJ.", indicates that my statement may not have been as clear as it should have been. Like I say I have been suffering from a dreadful cold. What I should have said was "I blame SMF software not DJ."

--In any case I want to thank you for correcting my stupid mistakes.

I wouldn't say your mistakes were "stupid".  We are all here to learn.  I have gained new appreciation for things I thought I knew, often pointed out by people very new.  We can all learn, and in such a rapidly moving field as electronics, we need to learn constantly or be left behind.