Searching for replacement MOSFET for old design

Hello Barry,

Thanks for the inquiry. Unfortunately, we have not been able to find any information on an EP2500 MOSFET made by TI. The best performing 200V, N-channel FETs from that era were probably something like the IRF640 or IRF630 from IR (now part of Infineon). Many vendors make a device under the same part number. These were optimized mainly for switch-mode power supplies (DC-DC and AC-DC converters). Sorry I can't be of more help but that's going back some time. Let me know if I can be of any further assistance.

Best Regards,

John Wallace

TI FET Applications

Hi Barry,

A colleague referred me to the link below which is a repository of TI data books. I have not had a chance to dig into it but maybe the part you're looking for is in one of these texts.

Thanks,

John

http://www.bitsavers.org/components/ti/_dataBooks/

Barry, I was intrigued by this question. I'm familiar with the TI archives at Southern Methodist University, and I really enjoy studying TI's history.

I came up with nothing on EP2500. Assuming it is a real TI part, the only guess I've got is that EP2500 is a preliminary number. I'm confident that P designates a power transistor.  I can't draw a conclusion from a negative, but I speculate EP meant something not generally distributed.  (Analogy with custom MOS part series numbers, and the TIXP40 example below.)

I am still digging up datasheets from the 70s, on the chance of finding a TI MOSFET in TO-220 for 200-300V, like IRF640 or 630. (John Wallace1)

Can you share a photo? Or any more context about the instrument?

Here's my reasoning:

1. In all the catalogs I checked, TI used either JEDEC numbers, or gradually introduced "House Numbers" like TIP02 for power transistors.  TIXP is found on a half-dozen parts.  (TI 1967-68 Semiconductor Components book, OEM price lists for 1966, 1975 and 1982)

2. No other manufacturer through 1978 had used the EP prefix, so I rule out its being a second-sourced part. (IC Master 1978)

3. I didn't yet find a functionally similar transistor in some TI Master Selection Guides.

4. The 1967-68 data book defines codes X and P for transistors. TIX and TIXP were development stage: available, but not sufficiently characterized. TIP designated a commercial power transistor. (The MOS division used P for preliminary.)

5. TIXP40 was in a former lobby display. The Smithsonian accepted (1987) the exhibits from TI's Dallas South Building lobby.  You can see the collection online. Caption: "TIXP designated an experimental design for one customer. This TIXP40 was registered in 1970 with JEDEC as 2N5940." I sent an email, and they courteously replied: they only have the captions, which TI employees wrote at the time.

 smithsonianchips.si.edu/.../t_159.htm

I wonder about that "one customer" description.  A half-dozen TIXP parts were cross-referenced in the front of the 1967-68 data book, as if they had been sold before.

6. Red herrings: none of the JEDEC numbers 250x were marketed in TO-220. Of the ones sold by TI:

1N2500 was a regulator (1966 price list, gone by 1975) Department code REG.
2N250x was a p-channel FET (1975 price list, 1978 IC Master) TI Department code was SM, for Silicon Mesa Planar (1975 price list) (very expensive at $20)
EPN2500 was a 1970s thermal printer from TI, utilizing the TMS2502 character pattern ROM.

See page 6303 (section 6 p 303)
www.bitsavers.org/.../1969_CC202_Preferred_Semiconductors_and_Components_from_Texas_Instruments.pdf

7. Wild guess: before 1966, I imagined EP as Epitaxial Planar (1966, Silicon Semiconductor Technology)  However, the price lists don't use this term. The 1975 price list gives department code SM as Silicon Mesa Planar. (not a prefix)

So that leaves me with the conclusion that EP2500 wasn't a generally available part number.  I believe the P is for Power, on its way to being a TIP house number. (Not Preliminary, which was how MOS used P.)  Since I came up with no published use of EP.  I still want to solve this mystery by looking for a functional match.

I enjoyed the challenge. It gave me some fresh ideas about TI's growth in the transistor market of the 60s and 70s.

A word about the TI Archive at Southern Methodist University: all are welcome to research there. But it is nowhere near comprehensive. Its real gems are the personal papers of TI engineers and managers. I observe that it doesn't reflect TI's worldwide presence (only Dallas and Houston.) And I'm sad to report their shelf space is full! (I wish they had room for the rest of C.B. Wilson's papers, which were nearly lost forever in a Dallas garage.)

Take care,
--Erik Olson
amateur of TI history

Hi Barry,

I suggest you and Erik become "friends" on this platform and communicate via private messages. Hover over Erik's name on the banner to send a private message and request friendship.

Thanks,

John