Spike suppression

My electronics is based on RAMPS but put together on a pad board using point to point wiring and some solder traces (connecting the pads). Recently, I've been experimenting with PID heater control in Sprinter. However, the rapid switching of the heater that results put very large spikes into my electronics which can give rise to spurious end stop activations - which makes for interesting prints.

The extruder heater is connected on one side to +12v and the other side is connected to the MOSFET. When the MOSFET is off, the MOSFET side of the heater measures 12v and when the MOSFET is on it is 0v (as you would expect).

+12v ------|HEATER|------MOSFET
                      ^
                      |

When you turn on PID control the power to the heater is controlled using PWM so the heater ends up being switched on and off very rapidly.

Measuring the voltage at the MOSFET side of the heater using an oscilloscope I can see that when the heater is turned off, the voltage goes from 0v (on) to 38v and then back to 12v (off). What is the best way to get rid of this spike? I've tried putting a diode across the heater and that improves the situation but the spike still goes up to 26v before returning to 12.

+12v ------|HEATER|------MOSFET
        |           |
        +---|<|-----+

Hi

A 14 to 16V suppressor- Diode will do the job. You will simply have to connect it between the 12V rail and ground. The polartity is not important, you can put it in both ways. The power rating of the diode should be at least a few watts. If a voltage higher than the rated voltage of the suppressor diode occurs it will switch into low resistance mode and absorb the power of the spike.

[en.wikipedia.org]

I've now tried one of those and it didn't make any difference. Maybe I chose the wrong one. What I got was P6KE16CA (breakdown voltage min:14.3, max:15.8).

CeeThreeEs Wrote:
-------------------------------------------------------
> Measuring the voltage at the MOSFET side of the
> heater using an oscilloscope I can see that when
> the heater is turned off, the voltage goes from 0v
> (on) to 38v and then back to 12v (off). What is
> the best way to get rid of this spike? I've tried
> putting a diode across the heater and that
> improves the situation but the spike still goes up
> to 26v before returning to 12.

The voltage spike should be limited by mosfet breakdown voltage. Most power mosfet's are specified in avalanche mode and should clamp the spike. Using lower-voltage mosfet could help.
Maybe RC snubber could help, but it can be quite tricky to get R and C values right. Maybe measuring inductance and some googling should povide reasonable values (probably tens of nF and around 100 ohms?)

The spike should be quite harmless unless current path is wrong - You should try using twisted wires to connect heater and check your ground layout (use single point where all grounds are conencted)

Thanks for all the suggestions. Turns out I lied and the TVS diode did do the trick - I just failed to connect it up right. All working much better now - spikes are much reduced so hopefully no more spurious end-stop activations mid-print.

Hi.
I know its long time since this topic...
But ant chance you can show how exactly did u wire the diode and what diode did u use.

I have the same problem now with printer stop printing mid print and dont know the reason trying to figure out the solution.

Tnx

A standard silicon rectifier diode is too slow suppress the spike effectively, but a Schottky diode will do the job. I would use a 1N5820 or similar. Be sure to connect it the right way round!

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].