Search

Title

Author

Article Body

Manager


Manager - AB7RG
Manager Notes

Sherlock Investigates: $25,000 Radio Fails Because of 2 Cent Plastic Part

Created by Paul Signorelli, W0RW on 2021-10-20

 

"Sherlock Investigates:

$25,000 Radio Fails Because of 2 ¢ Plastic Part”

 


   It was a dark and cloudy day when this defective radio was returned to the 221B Baker Street Laboratory for analysis. I had offered my services to the Ministry of Defense before to help with some of the more singular failures.


   Watson was unavailable to record these findings, so these writings are only from my sketchy notes. i hope they prove helpful to others faced with such an elusive and singular failure. This is a factual report, my monographs and case files contain fewer embellishments here than Dr. Watson’s.

 

   When the defective radio arrived it was wrapped in a plastic bag and the operator had attached some white mine-tape to the radio stating the following: "The field operator was on HP (High Power) when the radio first dropped its memory, (It zeroized itself). It had lost internal power for a few seconds and then came back on by itself but with no frequencies in left memory. (Was it a cosmic ray?).


   He re-entered the Command Frequency and the unit continued operating, but it tripped off again and dumped the memories again after 30 additional minutes of operation. Then after another 1 hour of operation it dropped out and would not turn on at all. The radio turns itself off with protection circuit command. It’s ‘Memory Palace’ was Erased!

 

   (Doesn’t this remind you of the Case of the Post Concussion Syndrome, PCS? Many years ago an electronic computation box lost its memory during vibration testing. The cause was a loose ‘tape wound core’ in a data transformer. When the core hit the bobbin case it lost its ‘sense’. They were used in space equipment because of their radiation resistance. This was before the appearance of Rad-Hard semiconductor memories).

 

   The Army Radio Technician confirmed the failure of the radio and sent it to the Depot. They have to keep track of sensitive items like this. This radio is a clandestine (spy radio), so none of the circuit boards have any silk screened reference designators, (e.g.: C1, R2, Q3, etc.).


   Some equipment’s’ like this have a small "red cross" painted on the case, to be used as an aiming point for rapid decommissioning. (Rapid decommissioning means Shoot and Run).

 

   These units have very rarely failed in the field, except for a few battery connections that were easily corrected. None of these were ever lost or decommissioned in the field.

 

   “You know my methods. There was not one of them which I did not apply to this inquiry, and it ended by my discovering the cause, but a very different one from which I had expected". Quotation from the ‘The Crooked Man’; A.C. Doyle, 1894.


   The first thing to do in failure analysis is to duplicate the failure without causing another one. A current limiting power supply was connected to the unit and found that it was drawing about 15mA, so it was not operational but didn't blow a fuse either.

 

   “It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts”. Quotation from ‘A Scandal in Bohemia’; A.C. Doyle, 1892.

 

   “I never guess, it is a shocking habit – Destructive to the logical faculty”. ‘How often have I said, that when you have eliminated the impossible, whatever remains, however improbable, must be the truth?’ ‘The Sign of Four’; A.C. Doyle.


   After the radio was opened the modular power supply was swapped with a ‘Known Good’ one and the unit began to work correctly for more than 2 hours. So we have a successful fault isolation to the Power Supply subassembly.


   The power supply module is small with only one switching transistor. The transistor had shorted. Secondary damaged was also found, (Secondary damage is damage caused by the primary failure). An integrated circuit and the transistor source resistors were also open.


   When the transistor (FET in a TO-220 case) was removed, It was noticed that the red plastic feed-through shoulder washer that is used to mount and insulate it from the heatsink was crushed. See Photo A. The transistor mounting screw was tight but the transistor tab was not fully seated against the chassis. If the tab is not flush it won’t conduct heat to the heatsink. That probably caused it to overheat and short out. The designer should have selected a shoulder washer that fit correctly. This assembly process is blind and the assembler could not see the miss-seating of the tab to the heatsink. See photos B and C.

 

 

 


   I therefore deduced that the CW duty cycle and long mission time reported by the field operator in high temperature environments probably caused the transistor to finally overheat. Previous missions may have only required a few minutes of SSB operation. That did not over stress the transistor and there was no hint of this latent defect.



   The failure could have been caused by other components, like tantalum capacitor scintillation, but that usually happens right when the radio is turned on. The scintillation usually either self-heals itself or causes a short circuit straight away.


   The unit probably lost its memory randomly several times because the protection circuits in the radio shut it down before the transistor could short out and destroyed itself. If this had not finally been a catastrophic failure the exact cause of the failure might have never been found. The memory is designed to erase when the power is removed for more than 30 seconds to keep unauthorized personnel from reading the operating frequencies.

 

   Continuing to the next level of analysis, the transistor was de-potted using hot M-Pyrol to expose the die. It was melted down from high current and thermal runaway.



   All the tantalum capacitors were also checked to make sure they were none reversed. There was a previous radio that failed after many hours of operation in the field because of a reversed tantalum capacitor.
It was reported in the "Case of the Three Orange Pips”, CQ Magazine, Nov. 19. P. 36-38.

 

   The radio has been returned to inventory and should live another long life. This case is closed. Other radios in inventory can be checked visually by sighting down the connector and inspecting the seating of the transistor, See photo D.

 


   This shows the value of real - time failure analysis during production. Failures are opportunities for reliability growth. This is another assembly process failure that was not detected during regular factory testing.

 

The Game is Afoot.
Sherlock (w0rw)

 

 

N8RKD2021-10-25
Sherlock Investigates: $25,000 Radio Fails Because of 2 Cent Plastic Part
I've seen transistors shorted to heatsinks in manufacturing when fasteners were overtorqued, crushing the insulating washer.
W0RW2021-10-22
Sherlock Investigates: $25,000 Radio Fails Because of 2 Cent Plastic Part
The radio is a green box. The failure happened in the Power Supply module.
You can see a great video of the 319 at:
https://www.youtube.com/watch?v=WCXb2M66fro
or just google 'PRC319' for more pictures.
Sherlock
KC6RWI2021-10-21
Sherlock Investigates: $25,000 Radio Fails Because of 2 Cent Plastic Part
Good read, odd case that radio is in, almost looks like a wooden shoe.
K0UA2021-10-20
Sherlock Investigates: $25,000 Radio Fails Because of 2 Cent Plastic Part
Excellent article. And excellent sleuthing.

73 James K0UA