Does the PD board Output 12v Stable?

My team is using a custom circuit that doesn’t have very much tolerance for voltages that aren’t exactly 12v. I know the battery jumps around a lot, but once it goes through the power distribution board, is the voltage exactly 12v, or should i look into other circuits that regulate voltage to 12v. Thanks.

Let me put it this way: There are motors, the controller, and everything else on the robot using electricity. A robot can go in with a full battery (which is a little bit over 12V) and come out with a battery almost dead (say around 5V-8V). The PD board’s job is to distribute that electricity to the rest of the robot, which means that it gets in (and out) whatever the battery is giving it. Don’t count on the voltage being very close to 12V for very long.

The 12V supply output is either 12V or the battery input, which ever is higher.

don’t you mean lower?

unless there’s some fancy circuitry, i don’t think 8 volts can magically become 12…

-Z

I do believe the 12V is a boost supply. So is the 24V line (Yes, there are 24 volts coming from the PD - this goes to the cRio.)

Doing conversions of voltage is common in circuitry. It’s much easier in an AC circuit, but entirely possible with some ingenuity with a DC circuit. Most ways, however, do include converting it to a high-frequency AC circuit and filtering.

The 5V line is not protected, though. I know that if battery drops below 5V, it will track that.

If your custom circuit requires a solid 12V, I recommend you add filtering and boosting to your circuit design. I do not believe it is legal to hook custom circuits up to anything except the PD through a 20 amp breaker. Don’t quote me on this, however.

Eric, aren’t you talking about the special 12V connection for the gaming adapter? As I understand it, that supply is regulated such that it will supply 12V even when the battery voltage is lower. However, the general-purpose 12V outputs (the ones attached to the individual Snap-Action breakers) are completely unregulated and will be at whatever voltage the battery is at, right? Assuming I’m correct with my understanding, the answer to the original poster’s question is that for the 12V supplies that you’re allowed to use for custom circuits, the answer is no, they are unregulated and therefore will not work for anything that is picky about 12V.

You are correct, that supply may only be used for the gaming adaptor.

The rest of the outputs are unregulated.

The 5V supply will only ever output 5V or 0V. If it can not provide a steady 5V, it gives up.

ZInventor - If you are interested, the schematics for the “fancy circuitry” are available in the competition manual section.

Is it too late to redesign your custom circuit to run off of 9V?

That would provide a stable threshold even if the battery voltage drops to say 10V as it could during a 2 minute match.

I’ve noticed that if a battery gets too low, the 24V booster voltage to the cRio will fail. I’m not sure of what the cutoff voltage is, but the robot will not run when it happens.

Keep the batteries fully charged this year when going to field.

Isolated DC-DC converters are readily available that will handel 9V - 18V inputs and provide a stable 12V output.

If you post the type of circuit and how much current is needed, I can suggest some possibilities.

Its a picoPSU-90, outputting 90 watts, for providing power to a motherboard we’ll be using to supplement processing power.

-jonathan

A little off-topic, but how are you passing the data between your motherboard and the cRio?

http://www.chiefdelphi.com/forums/showthread.php?t=73577

After this (mostly fruitless) discussion, I decided to just use serial with a max232 to reduce voltage. The motherboard most likely won’t need to receive input, and will send maybe 120 bytes a second back - note this is the current plan and as soon as we have a max232 we can check what kind of baud rate the cRIO can actually handle. I see no reason it can’t handle a bit a millisecond.

-jonathan

if you have any better ideas I’d be happy to hear them :o

Whoa! I was thinking a couple of watts.

For your application it makes more sense to use the M3-ATX which has a wide input range from 6V - 24V and produces a bit more power. Otherwise, you’ll need a 120Watt Plus DC-DC supply which will likely cost even more.

I suggest running M3-ATX on its own 20amp breaker and with all of the “automotive” features disabled so that it won’t shut down on you as the battery drops under 11V.

To answer a few things here. The battery voltage regularly will vary from 12 volts to as low as 3-4 volts depending on the load and state of charge. The 2009 game is unlikely to draw the battery so low during a match but in any design, expect at least a 5 volt variation in the unregulated supplies. IFI control systems used an eight volt trip point before switching to the backup battery and disabling the PWM outputs. Eric can correct me but I believe the PD 24 volt output will be maintained down to a sustained battery voltage of 4.5 volts as well as the 12v to 12 volt (for the wireless) and the five volt (for the camera) power supplies.

See R66 R67 and make sure you’re looking at the latest revs.

From the Q&A:

** The serial port is not available for use during the competition this year. Use of this port is not permitted.**

Also, I’m having a hard time imagining what you could possibly want from a secondary computer on board. The PowerPC processor in the cRIO is the most powerful CPU we’ve ever had at our disposal, by a factor of like 100. If it can’t do what you’re trying to do with an off-cRIO processor, I’d be surprised.

The 12V supply is only capable of 20 watts. Therefore, you can’t do this, regardless of the rules.

I am interested in what you are trying to accomplish. We might be able to help you find a better way.

In theory, one could connect a coprocessor to the control system via I2C and use the coprocessor to provide serial communication with some sensor. It would be a rather sneaky way to get around FIRST banning use of the serial port on the cRIO, but would be technically legal.

I suppose, though IMO, its alot of added electronics and stuff to talk to some sensor, and if it requires that kind of nonsense to talk to it, its probably more complicated of a sensor than you really require.

We are aware of this; we will wire the serial port to the gpio of a digital sidecar.

Our robot design has forced us to become something of a sniper - we should have the accuracy, and want additional speed in processing images. We may use a second camera to help with targeting.
The fact tha the old processor was 1/100th of the current one doesn’t mean the new one is blazing fast, it means the old one was very slow. :wink:
I don’t view ~450 mhz as very fast, particularly when we can only get image processing (at the lowest possible resolution) to run at about 20 fps, with nearly no other operations running simultaneously.

Do you mean 20 amps?
Watts = Amps x Voltage
20 = Amps x 12
Amps = 20/12 = 1.67

I’m pretty certain that the motors draw more than that.

-jonathan