Running more than 12v

The RAMPS wiki says you can run up to 32v into the shield as long as you don't install the diode. And Brent Crosby has shown that you can get some significant benefits from running 35v through the steppers on a Cupcake. [groups.google.com]

I was wondering what it would take to run more voltage through the steppers on a Mendel with RAMPS. Can you just put 35v into the power block instead of 12v?

To run more than 32 volts, you'd need to replace the Pololu driver boards with a chip that handles a much higher voltage.

The Pololu drivers (which of course drive the stepper motors) are the part that maxes out at 32 volts. The FET's (the only other active component on the board) should happily do 50 volts. This isn't useful to do though, as you're trying to drive the stepper voltage higher.

... with higher voltages the steppers can be faster stepped, so this mainly changes the dynamics of the machine.

I have driven steppers with 12 Volts and 1/8-microstepping and 35 Volts - with 35Volts i could get much higher max. speeds and better acceleration values.

My comercial steppers runs with 70 Volts or 325Volts ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Right, so, assuming you don't raise the voltage over the Pololu's max, can you just change the input voltage the board is getting? From the comment on the wiki it looks like you can put more than 12v into the board itself without any problems, and as long as you don't go over the max voltage of the stepper drivers the performance of the steppers will be improved (also limiting the current).

Are there any side effects of increasing the voltage going into the board?

Also, a related question, is the shield getting its logic voltage from the USB connection in the Arduino?

makeme Wrote:
-------------------------------------------------------
> Are there any side effects of increasing the
> voltage going into the board?

The FETs are the only other part driven off the standard 12V line. The FETs have a minimal on resistance, but it's still there. Any voltage drop across the FET's will be increased, so you may pull more power through the FET, which will mean you'll need to heatsink the FETs.

> Also, a related question, is the shield getting
> its logic voltage from the USB connection in the
> Arduino?

This is correct.

You could power the Adruino off another supply (it has a power socket on it), in case your USB isn't capable of the output. Most should be fine.

Okay, so you can't limit the current to the FETs like to the steppers? I didn't see a way to do that but I'm not particularly good at reading schematics either.

If I've got this right then the Arduino is supplying the logic voltage for everything because it's got the USB connection. So, you can run up to 30ish volts into the shield as long as you 1) limit the current going to the steppers and 2) heat sink the FETs (and probably the stepper drivers too).

I assume that means you're sending the same increased voltage to whatever else the shield is running, like the extruder heater and a heated bed. But you'd never run those without a closed-loop sensor, so it shouldn't really matter as it'll just be easier or harder for them to achieve and maintain a certain temperature.

Does all that pass muster with the EE guys?

The FETs are what drive the extruder heater and heated bed elements.

If you are using a higher voltage and want to reduce the current you use, you could probably increase the resistance of the Extruder heater and Heated bed elements (eg: using different resistors in an aluminium block heater, etc).

From the basic formulas P=VI and V=IR, we can get P=I²R or P=V²/R.

So lets assume the default voltage of 12 volts across the extruder heater element of 6 ohms, and we want to find the power (ie: how much it heats).

So for V=12 volts, R=6 ohms: 12²/6 = 24 watts

Now if you boost the voltage to 30 volts with the same setup...

So for V=30 volts, R=6 ohms: 30²/6 = 150 watts

... which is a lot more power than you need. You may find that during heating you overshoot the mark.

So If we want 24 watts, and we know the voltage is 30V, we have to change the equation around a bit to come up with this:

R=V²/P.

So for V=30 volts and P=24 watts: 30²/24 = 37.5 ohms.

This means that when using 30 volts, you could use a 37.5 ohm resistor and still get the same amount of heat output as a 6 ohm resistor at 12 volts.

That difference in power usage will make the bot more efficient, which means you may be able to use a smaller power supply. It will also avoid you drawing too much current and tripping the PTC fuse on the RAMPs circuit board. It will also reduce the amount of current flowing through the FETs, which means you may not need to heatsink them after all.

Note: I haven't allowed for the resistance of the FETs in these calculations, but the idea is much the same, just more difficult to work out, and therefore harder to show the workings in a forum post.

Hey, thanks a lot! I appreciate you taking the time to break it down for me.

As long as every component can handle a higher voltage, are there any negative side effects? It seems like it makes everything work better while also making things somewhat less likely to overheat. Power supplies are probably a bit pricier and/or more rare, but why isn't this a normal thing?

Pololu has a DC-DC converter that can step 12v up to 24v at 2A and is smaller than a quarter. [www.pololu.com]
It can also be fixed to output 30v (if you order a lot).

Field test: was accidentaly runing RAMPS at 40V for 1-3 seconds, (have forgoten re-set my lab power supply after heated bed test). One of pololu carier driver was shorted.. lately I found out there is shorted .22uF capacitor.. replaced by new one. Pololu is in the game again :-) However, not fancy as before.. i can't find a SMT one so was replaced by two 100n regular capacitors.

BTW. I made short info about building RAMPS 1.2 here [www.np-diskuze.cz]