HP transistor array 1821-0001 in 3456A

[rolkinas]

I want to share my experience replacing HP transistor array 1821-0001 in 3456A on A30 board.
some time ago this part failed in my 3456A. Replaced it with 5pcs of 2N3904 and it worked for some time. This suggestion was found somewhere in someone's blog. After some time unit started to be unreliable. Had to turn on oscilloscope which means it is serious repair
using 2N3904, timing on TP1 and TP2 on A30 board is way off. Same with new set of 2N3904. Looks like it was working just with help of devil.
Quick experiment showed that SS9018 transistors is almost perfect fit. SS9011 fits good also.
Pictures shows signal timing of original 1821-0001 (from working unit), 2N3904, 1821-0001, SS9011 and SS9018
sorry for different horizontal time base. Was on hurry.
Hope it saves someones day

[amspire]

I gather the 1821-0001 is a CA3046. Why aren't you replacing the array with a CA3046?

I have plenty of the smd version of the CA3046 (LM3046M). If you have trouble sourcing the chip, send me  message. I could post a few.

Richard

[rolkinas]

Thank you for advice! Cross reference of HP part numbers that I have, has no 1821-0001 mentioned, I thought that it is proprietary part. Now found new HP cross reference book.
Mouser has HFA3046BZ which looks to be current version
Anyway I have plenty of discrete transistors in storage it was fix on hurry. Currently SS9018 works just fine I will not replace them until next failure

[Kleinstein]

The circuit around the transistor array does not look like it needs matched transistors in the classical sense. So a replacement with single transistors is probably OK in this position.