Bad part can come from anywhere

WOPL Sniffer

Veteran and General Yakker
Joined
Aug 10, 2015
Messages
11,230
Location
Minnie-Soda
Tagline
Screw it
#1
I have to thank Lee, Glen,Joe and Don for assisting me with a problem which made me think I was a rookie...... I ordered the parts to build 5 DCP I had sitting here and after receiving all the parts from Mouser about 2 weeks ago, I started soldering them together. When finished, I installed a DCP in a fresh 700B WOPL I had just finished and when I applied power, there was no DELAY, the relays on the DCP immediately clicked in (think turn on thump).....

So, I monkeyed around and had to take the remaining transistors, still in the original packing and do some measurements... They didn't measure up.... I get an OPEN reading between the base and emitter even when forward biased...... Just because you got new parts, does NOT mean they will work.... Apparently......



IMG_4607.jpg IMG_4608.jpg
 

Skratch

Chief Journeyman
Joined
Dec 27, 2011
Messages
1,131
Location
Huntley Il.
Tagline
Life is short, so take your time
#2
I have had the same problem from Mouser, I had 1 ohm and 10 Ohm in the same bag and they were the same banding
 

WOPL Sniffer

Veteran and General Yakker
Joined
Aug 10, 2015
Messages
11,230
Location
Minnie-Soda
Tagline
Screw it
#3
Well, Joe figured it out. They have the alternate TO-92 pinout that switches the Emitter Base Collector pin out. Even though the bag says "On Semi", Joe says they are Fairchild and

"The pinout on the devices you got was E-C-B which is a commonly used alternate pinout for TO-92 transistors (to E-B-C which is more standard). If you swizzle the pin orientation, then things will work.

well, ya lern something new every day... The data sheet didn't show that the pins were "swizzled" and if they had, I may have figured this out a week or so ago...... I'm going to swizzle this one for MY amp and try it. I'll wait for the good ones to show up that I just ordered. There is a reason we pay a dollar for a good transistor and 4 cents for the shit.....




HUH???????
 

Gepetto

Administrator
Staff member
Joined
May 15, 2011
Messages
13,553
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#5
Last edited:

Gepetto

Administrator
Staff member
Joined
May 15, 2011
Messages
13,553
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#7
Well, Joe figured it out. They have the alternate TO-92 pinout that switches the Emitter Base Collector pin out. Even though the bag says "On Semi", Joe says they are Fairchild and

"The pinout on the devices you got was E-C-B which is a commonly used alternate pinout for TO-92 transistors (to E-B-C which is more standard). If you swizzle the pin orientation, then things will work.

well, ya lern something new every day... The data sheet didn't show that the pins were "swizzled" and if they had, I may have figured this out a week or so ago...... I'm going to swizzle this one for MY amp and try it. I'll wait for the good ones to show up that I just ordered. There is a reason we pay a dollar for a good transistor and 4 cents for the shit.....




HUH???????
When you see 512- in the beginning of the part number they are Fairchild
When you see 863- in the beginning of the part number they are ON Semi

ON Semi acquired Fairchild 2-3 years back and they are merging their part numbers. ON Semi end of life'd TO-92 transistors a couple years back when their packaging house decided they wanted to encapsulate higher value parts. ON Semi made the decision at that time to discontinue their TO-92 line up (which was really a shame).

Fairchild still has TO-92 packaging capabilities and that will probably mean a re-emergence of the full ON Semi line up over time (speculation on my part). In the meantime, be careful on the 512 parts until the QC processes sort out.

That is the back story.
 
Joined
Apr 1, 2015
Messages
133
Tagline
---
#8
Not quite exactly. I have run smack dab into this issue many many times.
I have STOPPED using part numbers that can generate this confusion.
Too darn many people on Karma got caught by it and sucked up my time troubleshooting...

There are two variants of the transistor and THREE parts numbers from Fairchild / OnSemi stocked by Mouser.com
512-2N6517TA

512-2N6517BU


512-2N6517CTA

Downloading the datasheet DIRECTLY from mouser, I looked closely and the third line in the sheet said:
Suffix “-C” means Center Collector (1. Emitter 2. Collector 3. Base)
thus the 512-2n6517bu is an EBC layout transistor shipped loose (bu = bulk)
and the 512-2n6517ta is an EBC layout transistor shipped on a "tape".
while
the 512-2n6517cta is an ECB layout transistor shipped on a "tape".
ALL according to the data sheet I downloaded. The "default" layout specified by the sheet is EBC.



The EBC is traditionally on "american" transistors i.e. 2n numbers, while the ECB is traditionally a "japanese" layout.
whatever the history
the part numbers DO correspond predictably with the data sheet.

It could be WORSE!!! They could (and DO) on others!!) have gain (Hfe) codes before the packaging specification
 
Last edited:

Gepetto

Administrator
Staff member
Joined
May 15, 2011
Messages
13,553
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#9
Not quite exactly. I have run smack dab into this issue many many times.
I have STOPPED using part numbers that can generate this confusion.
Too darn many people on Karma got caught by it and sucked up my time troubleshooting...

There are two variants of the transistor and THREE parts numbers from Fairchild / OnSemi stocked by Mouser.com
512-2N6517TA

512-2N6517BU


512-2N6517CTA

Downloading the datasheet DIRECTLY from mouser, I looked closely and the third line in the sheet said:


thus the 512-2n6517bu is an EBC layout transistor shipped loose (bu = bulk)
and the 512-2n6517ta is an EBC layout transistor shipped on a "tape".
while
the 512-2n6517cta is an ECB layout transistor shipped on a "tape".
ALL according to the data sheet I downloaded. The "default" layout specified by the sheet is EBC.



The EBC is traditionally on "american" transistors i.e. 2n numbers, while the ECB is traditionally a "japanese" layout.
whatever the history
the part numbers DO correspond predictably with the data sheet.

It could be WORSE!!! They could (and DO) on others!!) have gain (Hfe) codes before the packaging specification
And to add further to any confusion, ON only made the EBC configuration and seems they use the CTA suffix for formed (splayed leads) tape and reel, while Fairchild made both pin configurations and used the C to indicated the alternate pinout config.
 
Last edited:

laatsch55

Administrator,
Staff member
Joined
Jan 14, 2011
Messages
74,244
Location
Gillette, Wyo.
Tagline
Halfbiass...Electron Herder and Backass Woof
#10
Were the ECB configurations to replace the Japanese transistors , or are there design considerations for that pinout??
 

Gepetto

Administrator
Staff member
Joined
May 15, 2011
Messages
13,553
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#11
Were the ECB configurations to replace the Japanese transistors , or are there design considerations for that pinout??
Generally speaking, and I say generally, the higher collector current TO-92 devices had the center collector for better heat/power dissipation purposes. Much like the TO-220 have center collector and tab connection for that same reason.
 

WOPL Sniffer

Veteran and General Yakker
Joined
Aug 10, 2015
Messages
11,230
Location
Minnie-Soda
Tagline
Screw it
#13
In MY book, they should not carry the same basic part number if they are not the same.

The Bu is bulk
The CT would be tape
And CTA Cut Tape

But they carry the same basic part number so the suffix should be the order delivery method. They should not rely on "small print" on the data sheet and not actually show the different pin out.
 

Gepetto

Administrator
Staff member
Joined
May 15, 2011
Messages
13,553
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#14
In MY book, they should not carry the same basic part number if they are not the same.

The Bu is bulk
The CT would be tape
And CTA Cut Tape

But they carry the same basic part number so the suffix should be the order delivery method. They should not rely on "small print" on the data sheet and not actually show the different pin out.
Agree Perry
They should have a different JEDEC part number for that device.
 

wattsabundant

Chief Journeyman
Joined
Apr 26, 2011
Messages
848
Location
Central Ohio
#16
I ran into this some time ago. The transistors sat around for several months until I needed them. At that point it was too late to return them. I tossed them in the trash.
 

Gepetto

Administrator
Staff member
Joined
May 15, 2011
Messages
13,553
Location
Sterling, MA
Tagline
Old 'Arn Enthusiast
#17
Use the MPSA42 or ZTX857STZ and you will not run into this problem again. There is no alternate pinout for these devices thus no confusion.

The ZTX857 is a much more robust device to boot.
 
Top