Diodes dying on me

Thread Starter

Edmunds

Joined Sep 27, 2010
85
Hi all,

I have a problem with diodes constantly blowing out on my two Velleman MK138 thermostat kits. The thermostats are part of a 3D printer setup and are powered from the same PSU as the stepper motor electronics. My suspicion is, this is because of the current and possibly voltage surge during "power on". The schematics says, the input current should be 100mA. My guess is, the only purpose of the diode is reverse polarity protection in which case I could just jumper it. On the other hand, possibly the diode D2 failing is what protects the IC from disaster.

How can I limit the current to 100mA? Any other possible causes or solutions?

Schematics of the thermostat attached below.

Thank you all for your input.

Edmunds
 

Attachments

ericgibbs

Joined Jan 29, 2010
18,766
Hi Ed,
IMHO the 1N4148 is not suitable, I would replace with say a 1N4001

The 100mA refers to the max current when operating, the C4 initial charge current will take out the 1N4148.
E
 
Last edited:

ErnieM

Joined Apr 24, 2011
8,377
A 1N4148 is described as a "small signal" diode, meaning it has very small genitalia when dealing with current.

Now at every turn-on there is very little to limit the current into C4. Additionally, if your motor is making hash on a common power supply that hash may make a signal to charge C4 so there is always some current keeping this device hot.

I hate that diode in that place. I would switch to a power diode such as a 1N4001-1N4007 which is a 1 amp power device. While you are at it, change D1 to the same.

I would expect your problems go away with that change.
 

Thread Starter

Edmunds

Joined Sep 27, 2010
85
Thank you both for your input and explanation. Changed both D2s to 1N4007s and one of the D1s as well, as I thought that might be the cause for a problem that it was not after all. Now both boards are fine again and let's see how long they last. It is funny though, that such a popular board that has probably been sold in many thousands of kits, has this trivial error in its design.

Edmunds
 

ericgibbs

Joined Jan 29, 2010
18,766
Thank you both for your input and explanation. Changed both D2s to 1N4007s and one of the D1s as well, as I thought that might be the cause for a problem that it was not after all. Now both boards are fine again and let's see how long they last. It is funny though, that such a popular board that has probably been sold in many thousands of kits, has this trivial error in its design.

Edmunds
hi Ed,
The output impedance of the Users power supply will be a significant factor.

If the impedance of the supply is 'highish' it will not be able to supply the very high initial charge current for C4.

Eric
 

crutschow

Joined Mar 14, 2008
34,283
hi Ed,
The output impedance of the Users power supply will be a significant factor.

If the impedance of the supply is 'highish' it will not be able to supply the very high initial charge current for C4.
Good point. That kit is likely typically powered by a small power supply, such as a wall wort, which could supply only a limited surge current. The stepper motor power supply obviously would be able to supply a much larger current. Sometimes a bad design only has a problem in unusual circumstances.
 

dougc314

Joined Dec 20, 2013
38
Have you had any D1 fail? - They shouldn't be. That diode is there to catch the inductive kick of the relay turning off. When T1 turns off the inductance of the relay coil has stored ~35 mA, which will continue to flow until the current is dumped. The collector will fly up in voltage and D1 will start to conduct when it gets above VCC by ~.6 volts. However the coil only has about 35 mA stored which should not hurt the 1n4148. The entire circuit will draw about 40 mA when the relay is energized and less than 10 mA when the relay's off. However with my printers the 12V supply can deliver several amperes because it runs both the steppers and the heaters, so inrush could easily pop a 1n4148 for D2.
 
Top