Neil Kurzman said that , “I can assume you have an R/C reset on the 8052?”
Thanks Neil… I’m already using it.
It can’t solve this situation… Because proses ordinary is;
1-R/C Reset
2-8051 Start up ( It’s writing ports to logic-1)
3-My program…
Neil correctly guessed that you had the worst possible reset circuit.
RC reset is horrible and may lead to flash loss. Also with a proper supervisor, you may even be able - while having the pulse longer than the minimum the chip requiers - to have a reset so short that the relays do not click.


Thans Richard. I know them.
8051 I/O maximum output current is 15-20mA (approximitly)
NO!, NO!, NO! some 8051 derivatives have I/O maximum sink current is 15-20mA (approximitly) however mayny derivatives will be specified with less or - a few - with more.

I won’t calculate at length. But, I/O pin output current through to Base pin is approximitly 1,9mA.
However, 8051 can supprot 15-20mA. (approx. 10 times more)
Due to that reason, current don’t have to flow outside to I/O pin.
If I could read what you thought you wrote, maybe I would understand the above.

BTW, yes at first time low-active design was significat.
ULN2003 etc. chips are used. But, some problems are noticed.
Now, Hi-Active usage has started lately. UDN2981 etc..
if you are already using such a chip to drive your relays, connect the suipervisor output to the enable pin and you are home free,

I think, I should use PIC series microcontroller for this kind of circuits….
how is "inactive open" ('input') better than "inactive high" in a previous post you babbled about an open solder joint, what is the difference betheen an open pin and an open solder joint.

Hundereds, if not thousands, have made succesful designs using the '51 to drive relays, it is remarkable that, after dacades,someone pop up that cant.

Erik

PS: from the jokers dictionary relay: recycling old girlfriends