Power protection on PSoC with regulated VCC

Hi Enrico,

I don't think using a zener between VCC and GND will achieve what you want it to achieve. I'd suggest you need your VCC external input to the PSoC to be supplied through a diode with low Vf such that it it reverse biased when the board power is off and the VCC from the PSoC will not then power the board. The issue would be whether your generated VCC - Vf of the diode would be high enough as an input to the PSoC. I've not looked at the specs of the PSoC so I don't know what it's valid input range is. Possibly another way to do it would be to have a MOSFET to switch on the VCC to the PSoC when the board it powered. It'll be off and high impedance otherwise. It's early and I still have low caffeine levels so I will have a bit more of a think on this and if I come up with a better answer I will let you know

Best Regards,

Rachael

Hi Enrico,

It is an interesting question, because it would apply to many scenarios where people want to use the Cypress board.

I think there are a couple of things that could be done simultaneously, one is that a diode would be good to protect the PC in case the 12V-derived 5V supply is too high, and looking at the PSoC board, there are a couple of places it could be fitted; one is to cut the VCC connection at the snap-off portion and place the diode in series there, or another way is to remove the fuse F1 on the board and replace it with a diode. This ON-Semi MBRM110L diodeON-Semi MBRM110L diode should fit, and has very low Vf so hopefully it should have no negative impact.

This way the PC (through the USB connection) will not accidentally be supplied with 5V derived from the 12V source.

The other issue as you say is that if your 12V-derived 5V supply is powered off, then the PC (through the USB) could end up powering some unexpected circuitry. Maybe a quick fix is to place a small resistance (e.g. 10 ohm) in the 5V supply line from the PSoC board, e.g. at SV1 pin 1 in your schematic, so that the amount of current that can be supplied from the PC USB connection is limited. With the PC USB disconnected, the 10 ohms will still allow the PSoC board to be powered from the 12V-derived source because the PSoC board consumes little current.

What do you think?

Thank you Rachel!

you are addressing me on the right way despite the low level of caffeine your grey cells seems working almost fine

Enrico

Hello Shabaz, thank you for your precious suggestions and the very complete explanation.

Respect the first part of your post, you say

one is that a diode would be good to protect the PC in case the 12V-derived 5V supply is too high, and looking at the PSoC board, there are a couple of places it could be fitted; one is to cut the VCC connection at the snap-off portion and place the diode in series there, or another way is to remove the fuse F1 on the board and replace it with a diode. This ON-Semi MBRM110L diode should fit, and has very low Vf so hopefully it should have no negative impact.

I will adopt the first option for two reasons; the first is that it is almost easy to implement without changing too much the schematics. But the most important thing is that as an external circuit will exist for sure I think it is more clear to complete the circuitry around the PSoC 4200 kit instead of changing it; I plan to setup about 50 of these boards and it is better to put the small kit in place instead of replicating the same fuse replacement to all the devices. There is also the consideration that if in future this small dedicated production will become a sort of product for some reason the PSoC kit will be replaced by the bare chip with the minimal electronics around it just to support this particular features (with a meaningful reduction of the entire board size).

About the limiting resistor it sounds also a good yet simple solution. Also considering that there is not a real damage but the USB programmer wont work at all. If I am not wrong the resulting power terminal should be something like the image below.

Enrico

Hi Enrico,

I drew the sketch below to show my current understanding, I'm not sure if I've completely understood it or not.

Does it look correct?

I've labelled all the potential areas where things could be modified with the letters A, B, C...F.

Regarding the resistor, I think it has to go at point D or E preferably (or points A, B or C less preferably), this way any power from the PC does not end up supplying the Stepper Driver board or any other circuitry attached in that area, because it will only provide limited current. As a result the PC should not become unstable.

For the diode, this is just an idea to prevent the 12V supply generated 5V from sending power to the PC by accident. If the diode (shown at position B in the diagram) is in position A,B or C, then when both the PC USB is connected and the 12V supply is connected, there is no risk of the PC getting damaged by the 12V supply generated 5V. If the diode was in position D or E or F then with the orientation as shown (i.e. Anode on the left) in my diagram, it would not be possible for the 12V generated 5V supply to power the microcontroller, so this would not be good. With the diode rotated 180 degrees (anode on the right), then it would work for any position A,B,C,D or E, I think it is a good idea, it will reduce the supply slightly and hopefully therefore not damage the PC USB port, although the protection is less than the anode-on-the-left confiigurations which have the limitation that they can only be placed at A,B or C.

So, in summary, I think diode and resistor can be at the same location (i.e. position D or E) with the anode-on-right.

Hello Shabaz, what was not 100% clear was the presence of the diode, you mean in series to the USB. So it is not correct in my previous design but it is useless in the usage context of this circuit. The only real problem is when the USB is connected to the kit, it tries to power the board, if by accident you have not disconnected the point D. So this will be solved by the resistor.

The opposite will never happen because when the stepper driver controller is engage there can't be any PC to USB connected at all.

Enrico

Hi Enrico,

Ah, I see! Perfect. In that case just the resistor is sufficient I agree.

Enrico Miglino wrote:

 

what was not 100% clear was the presence of the diode, you mean in series to the USB. So it is not correct in my previous design but it is useless in the usage context of this circuit. The only real problem is when the USB is connected to the kit, it tries to power the board

The point of the diode shabaz had suggested was to protect the USB on your PC in case of a fault on your main board power supply giving greater than 5V output. Other solutions which were just on the main board might involve some arrangement to clamp the voltage to the PSoC so it can't go above say 5.5V. His suggestion of the 10R resistor was to overcome your original issue and this will probably be fine but if you are looking to take this to production then you might want to consider all the options on this to make sure it's reliable.

Best Regards,

Rachael

Hi Rachel,

thank you for the clarification. My initial supposition was based on the use of a zener (applied in other circuits) so I continue to focus my attention over the power side of the PSoC.

In the case this board goes in production as I mentioned it will be part of a more complex assembly and no one kit will be used. In this second case the USB problem does not exists at all as while the kit is very useful to program with the USB and a bootlaoder an ideal production version will not include the bootloader and the chip will be programmed via a programmer.

Enrico