Help finding specs for transistor - RCA K68 = Motorola K68MEX

[SilverSolder]

I have a few of these transistors that I got at a surplus store, but this one is driving me crazy...  not yielding to my search efforts with Google or anything else I could think of.

Does anyone know what this transistor is equivalent to,  or even have a spec, or any pointers at all to help figure this out?

The transistor tester says it is an NPN transistor, with a Beta of 32 and Vbe of 559mV...

[duak]

I think the U8435 is the date code: week 35 of 1984.  I think K68 is the part number.

Sometimes an obvious prefix is left out.  A Japanese-ish 2SK68 doesn't bring up anything useful.  Perhaps a European prefix?  Pro-electron doesn't seem to use 'K'

RCA used to sell replacement parts with an SK prefix.  I remember semiconductors started with SK30 but SK3068 turns up a selenium rectifier.

[Benta]

Custom marked. Very common back then, your chances of finding what it is are extremely slim.

[amyk]

Try looking through 1980s RCA databooks.

[SilverSolder]

Thank you for the comments so far.

I have found an RCA book "1986 RCA Bipolar Power Devices" but there is nothing that relates back to any of the numbers printed on the transistor.

Wonder if it is possible to measure a few basic parameters of a specimen and see if something in the catalog matches... ?

[SilverSolder]

Some progress!

In the same bag of transistors, there were two Motorola transistors marked Motorola (logo) K68MEX, datecoded '85.

Presumably, these are equivalent to the RCA K68 since they (a) were found in the same bag, and (b) have similar part numbers (K68)

Sadly, searching for the Motorola part number yielded no results either...  but at least it is clear that the part was made by more than one manufacturer.  Is it perhaps less likely that it was a custom marked part if it was available from two different manufacturers?

[duak]

A long shot is to find a company name that starts with 'K' that might have used something like this in the 70s to 80s.  It was big enough to get two semi manufacturers to bother to print something on the parts but small enough that they didn't need too many types of transistors.

I looked up  Keithley and their part numbering scheme doesn't seem to use the letter K.

[amyk]

Someone found K75MEX in this thread:

https://www.eevblog.com/forum/repair/motorola-u8344-u8338-who-knows-these-guys/

[SilverSolder]

My best guess at this point is that these K series transistors are a tighter-spec variant of a standard part.

Detective work so far:

We can rule out some candidates right away -
1) It is not an extra high amperage part (the pins are 40 mil diamter, not 60 mil like the high current TO3 housings)
2) It is not a very high voltage part (the isolator around the pins is not extra wide, like it is on high voltage TO3 housings)

This still leaves quite a few to choose from.

Some measurements of K68:

Vceo:  >310V  (3.6uA leakage at 310V, the max of the power supply)
Hfe: 32 (measured with simple transistor tester)
Cobo: 900pf (measured with trans. tester. Divide by about 4 to get capacitance with bias - say around 225pF)


One possible suspect for the standard part is 2N6688.  It has a kind of poetic link to the part number K68...

Specs for 1980's RCA 2N6688

Vceo(sus): min 200V
Hfe: min. 25
Cobo - max 300pF


It doesn't seem completely crazy that K68 could be a selected version of 2N6688.


Any ideas for further tests that could be done to help confirm or disprove the null hypothesis that K68 = 2N6688 ?

[amyk]

If K68 is 2N6688, then is K75 2N6758 ? A 2N6758 is also in a TO-3 case but unfortunately it's a N MOSFET, so that theory is dead... but you can check some of the other parameters like Vbe and Vce(sat) to see if they match.

Also, I think "MEX" is just Mexico, where they were made.

[SilverSolder]

If K68 is 2N6688, then is K75 2N6758 ? A 2N6758 is also in a TO-3 case but unfortunately it's a N MOSFET, so that theory is dead... but you can check some of the other parameters like Vbe and Vce(sat) to see if they match.

Also, I think "MEX" is just Mexico, where they were made.

Testing Vce(sat) is a bit of a challenge, i.e. finding a power supply capable of 20A - 30A...   Maybe I can use a car battery or something...

Or perhaps a pulse load would work,  out of a suitably big capacitor?

[vk6zgo]

Some progress!

In the same bag of transistors, there were two Motorola transistors marked Motorola (logo) K68MEX, datecoded '85.

Presumably, these are equivalent to the RCA K68 since they (a) were found in the same bag, and (b) have similar part numbers (K68)

Sadly, searching for the Motorola part number yielded no results either...  but at least it is clear that the part was made by more than one manufacturer.  Is it perhaps less likely that it was a custom marked part if it was available from two different manufacturers?

I would guess "MEX" meant the devices were manufactured in Mexico, as I have seen that suffix on other more well known types.
We often received mixed bags of devices, some with "MEX" on them, & some without.

Very early RCA devices used the prefix "CK", maybe an early Silicon power device, with the "C" omitted?

I know most early power devices were germanium PNP, but there were definitely some odd types out there.

Some very early "transistor radios" had NPN germanium small signal devices, but they were pretty much swamped by the flood of early PNP germaniums in the "OC xx"series & their "2Nxxx" & "2Sxxxx" counterparts.

Philips & Mullard actually made a few real "bipolar transistors" (they had interchangeable emitter & collectors for "special applications", the details of which are lost in the mists of time.)

[SilverSolder]

Some progress!

In the same bag of transistors, there were two Motorola transistors marked Motorola (logo) K68MEX, datecoded '85.

Presumably, these are equivalent to the RCA K68 since they (a) were found in the same bag, and (b) have similar part numbers (K68)

Sadly, searching for the Motorola part number yielded no results either...  but at least it is clear that the part was made by more than one manufacturer.  Is it perhaps less likely that it was a custom marked part if it was available from two different manufacturers?

I would guess "MEX" meant the devices were manufactured in Mexico, as I have seen that suffix on other more well known types.
We often received mixed bags of devices, some with "MEX" on them, & some without.

Very early RCA devices used the prefix "CK", maybe an early Silicon power device, with the "C" omitted?

I know most early power devices were germanium PNP, but there were definitely some odd types out there.

Some very early "transistor radios" had NPN germanium small signal devices, but they were pretty much swamped by the flood of early PNP germaniums in the "OC xx"series & their "2Nxxx" & "2Sxxxx" counterparts.

Philips & Mullard actually made a few real "bipolar transistors" (they had interchangeable emitter & collectors for "special applications", the details of which are lost in the mists of time.)

Looks like we are talking 1980's here, based on the date codes on these devices.  The Vbe measures as 559mV, which points to silicon rather than germanium.

I'm still thinking hard about how to generate a 20A test current to measure Vce(sat)!   

I can get to 15A with my beefiest supply...   might as well be 1.5A  for all the good that does...

[Laser Fish]

Hi, I know I might be a little late to the table but I've come across the same transistors. Hence; your form. However I was fortunate enough to come across a RCA 2N6308 in the same lot. I don't know enough about these parts to tell if it's ~similar / match, but I thought someone on this form could tell. LMK also typical applications. Thanks, LF