Hello. First time poster.
I have a question about power sequencing in switching power supplies. I want to use a modern switching power supply to power a vintage Intel 8080 microprocessor circuit. The target circuit requires three voltages:
+12V @ 350 mA
+5V @ 1.3 A
-10V @ 200 mA (which is converted on-board to the -5V substrate bias voltage needed by the 8080.)
Each of these three voltage inputs is buffered on-board with a 22uF capacitor to ground.
I would like to power this board with a multiple voltage switching power supply (Mean Well RT-50B) with the following specs:
+12V @ 2.0 A
+5V @ 4.0 A
-12V @ 0.5 A (The 79M05 -5V voltage regulator on the 8080 board has a wide operating range with a maximum of -35V, so the higher voltage should be safe.)
From the standpoint of providing adaquate power, this supply appears to fit the bill. However, I found a warning on the 8080 Wikipedia page that I did not find in any of the Intel datasheets or literature:
"The -5 V power supply ... must be the first power source connected and the last disconnected, otherwise the processor will be damaged ... The +12 V power supply ... must be the last connected and first disconnected power source."
In researching this requirement, I found this in Steve Ciarcia's "Ciarcia's Circuit Cellar, Volume 3" (available for free in Google Books), p. 109, with regard to 4116 RAM:
"The most important thing to remember when designing any computer that uses 4116s is that the power-supply voltages have to be turned on and off in sequence. To keep from blowing the 4116 on power-up, the -5V supply must be turned on before the +5V and +12V supplies. On power-down, the -5V has to remain on while the +5 and +12 are removed. In lower-current power supplies (such as you will probably use), the sequencing can be accomplished through the time-constants of the power supply itself. This technique is used in the TRS-80. By giving the -5V section a very fast time constant compared to the other two supplies, it appears to come on first. On power-down, the sequence is reversed. Because the -5V has such a low-current draw on it, it will stay up long after the other voltages have dropped."
and on page 103:
"Power-supply sequencing is important because many power supplies overshoot their rated voltages when they are turned on. If VSS (-5V) is not turned on and VDD (+12V) over-shoots to more than +15V, the [4116] chip will blow. Applying VSS first provides an extra margin to prevent device destruction."
Now back to the power supply I want to use:
Although I have no schematic of the power supply, after spending an evening reading about the secondaries of switching power supplies, it seems intuitive that in order for the +12V and +5V outputs to source 24W and 20W respectively, they would have to have larger L and/or C values in their output sections than the -12V output, rated for 6W, and thus should have longer time constants.
Thus I hypothesize that this power supply will be safe for this circuit because the -12V output is expected to come up faster than the +5V and +12V supplies, thus applying the proper power-on sequence, and the low negative voltage power consumption will make the -5V source outlast the +5V/+12V sources on power-off.
But I ask before I test, out of caution. I don't want to let out the 40 year old magic smoke. I'd appreciate any comments and tips from those with a better understanding of switching power supplies.
Thanks,
John Schulien
Moderator Scott Wang : Please do not post any info about Email, it may cause email spam problem.
I have a question about power sequencing in switching power supplies. I want to use a modern switching power supply to power a vintage Intel 8080 microprocessor circuit. The target circuit requires three voltages:
+12V @ 350 mA
+5V @ 1.3 A
-10V @ 200 mA (which is converted on-board to the -5V substrate bias voltage needed by the 8080.)
Each of these three voltage inputs is buffered on-board with a 22uF capacitor to ground.
I would like to power this board with a multiple voltage switching power supply (Mean Well RT-50B) with the following specs:
+12V @ 2.0 A
+5V @ 4.0 A
-12V @ 0.5 A (The 79M05 -5V voltage regulator on the 8080 board has a wide operating range with a maximum of -35V, so the higher voltage should be safe.)
From the standpoint of providing adaquate power, this supply appears to fit the bill. However, I found a warning on the 8080 Wikipedia page that I did not find in any of the Intel datasheets or literature:
"The -5 V power supply ... must be the first power source connected and the last disconnected, otherwise the processor will be damaged ... The +12 V power supply ... must be the last connected and first disconnected power source."
In researching this requirement, I found this in Steve Ciarcia's "Ciarcia's Circuit Cellar, Volume 3" (available for free in Google Books), p. 109, with regard to 4116 RAM:
"The most important thing to remember when designing any computer that uses 4116s is that the power-supply voltages have to be turned on and off in sequence. To keep from blowing the 4116 on power-up, the -5V supply must be turned on before the +5V and +12V supplies. On power-down, the -5V has to remain on while the +5 and +12 are removed. In lower-current power supplies (such as you will probably use), the sequencing can be accomplished through the time-constants of the power supply itself. This technique is used in the TRS-80. By giving the -5V section a very fast time constant compared to the other two supplies, it appears to come on first. On power-down, the sequence is reversed. Because the -5V has such a low-current draw on it, it will stay up long after the other voltages have dropped."
and on page 103:
"Power-supply sequencing is important because many power supplies overshoot their rated voltages when they are turned on. If VSS (-5V) is not turned on and VDD (+12V) over-shoots to more than +15V, the [4116] chip will blow. Applying VSS first provides an extra margin to prevent device destruction."
Now back to the power supply I want to use:
Although I have no schematic of the power supply, after spending an evening reading about the secondaries of switching power supplies, it seems intuitive that in order for the +12V and +5V outputs to source 24W and 20W respectively, they would have to have larger L and/or C values in their output sections than the -12V output, rated for 6W, and thus should have longer time constants.
Thus I hypothesize that this power supply will be safe for this circuit because the -12V output is expected to come up faster than the +5V and +12V supplies, thus applying the proper power-on sequence, and the low negative voltage power consumption will make the -5V source outlast the +5V/+12V sources on power-off.
But I ask before I test, out of caution. I don't want to let out the 40 year old magic smoke. I'd appreciate any comments and tips from those with a better understanding of switching power supplies.
Thanks,
John Schulien
Moderator Scott Wang : Please do not post any info about Email, it may cause email spam problem.
