Author Topic: TMC5160 keeps dying on 24V input  (Read 1468 times)

0 Members and 1 Guest are viewing this topic.

Offline semir-tTopic starter

  • Regular Contributor
  • *
  • Posts: 56
  • Country: ba
TMC5160 keeps dying on 24V input
« on: May 29, 2023, 05:51:35 am »
Hello all,

I have custom board with 2 TMC5160 stepper motor drivers. The schematic is based on TMC5160 SilentStepStick which by datasheet states that the input voltage is in range from 10-30V. Attached is my schematic.

Board works great on 12V and even up to 18V as far as I can see it. But, as soon as I connect it to 24V (which is requirement) it doesn't work any more and the TMC5160 gets a little bit hotter than usual. After that even if I reduce the voltage and try it on 12V it doesn't work any more.

Did anyone else had this problem and any idea what could be wrong on my board ?
 

Offline moffy

  • Super Contributor
  • ***
  • Posts: 1682
  • Country: au
Re: TMC5160 keeps dying on 24V input
« Reply #1 on: May 29, 2023, 06:36:23 am »
1. You could try powering VSA from a supply separate from the motor supply, to see if noise from the motor supply is affecting it.
2. On your schematic I didn't see a 100nF capacitor, should be ceramic preferably surface mount, connected to the 24V supply or VS near the chip. It is shown on the basic application, you have 10u & 1u but lower value caps have higher frequency performance. Proximity to the chip is usually important also.
Just some thoughts.
 

Offline semir-tTopic starter

  • Regular Contributor
  • *
  • Posts: 56
  • Country: ba
Re: TMC5160 keeps dying on 24V input
« Reply #2 on: May 29, 2023, 06:51:42 am »
Thanks for the reply,

1. I noticed that after the boards were produced. This is something that I am thinking for the next spin of the board but at the moment I am trying to figure out what could be causing this. For me personally, internal LDO should be fine with 24V as the chip is graded to lot higher. I also have mounted heat-sink.
2. The 10u and 1u are close to the chip. To be honest based on the clip () I didn't place 100nF but 10u and 1u. Prof. Eric explains on 55:55 about this. It is not a problem for me to add the 100nF as well and test it but I am afraid that if I connect it it will burn my chip and I have only 3 left.

Best regards,
 

Offline Swainster

  • Frequent Contributor
  • **
  • Posts: 258
  • Country: sg
Re: TMC5160 keeps dying on 24V input
« Reply #3 on: May 29, 2023, 07:20:47 am »
I couldn't see the ePad on your schematic
 

Offline semir-tTopic starter

  • Regular Contributor
  • *
  • Posts: 56
  • Country: ba
Re: TMC5160 keeps dying on 24V input
« Reply #4 on: May 29, 2023, 07:22:19 am »
ePad is connected to GND. This is how I created the footprint.
 

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 6459
  • Country: nl
Re: TMC5160 keeps dying on 24V input
« Reply #5 on: May 29, 2023, 07:37:42 am »
Didnt check your schematic but is the TMC dead or the external fets?
If the fets check the max GS voltage, often this is 12 or 20V max, so you need an voltage divider in that case to the gate. Just a thought.
 

Offline semir-tTopic starter

  • Regular Contributor
  • *
  • Posts: 56
  • Country: ba
Re: TMC5160 keeps dying on 24V input
« Reply #6 on: May 29, 2023, 07:44:46 am »
It looks like that the TMC is dead because when I replace it with the new one the board works again on 12V.

My though is that if it were the FETs when I replace the TMC the board wouldn't work.
 

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 6459
  • Country: nl
Re: TMC5160 keeps dying on 24V input
« Reply #7 on: May 29, 2023, 07:57:38 am »
True.
 

Offline PCB.Wiz

  • Super Contributor
  • ***
  • Posts: 1473
  • Country: au
Re: TMC5160 keeps dying on 24V input
« Reply #8 on: May 29, 2023, 08:52:44 am »
.... But, as soon as I connect it to 24V (which is requirement) it doesn't work any more and the TMC5160 gets a little bit hotter than usual...

How exactly do you 'connect to 24V' ?
It's a known issue that hot plug in of a pre-charged plug pack/cable, can overshoot with the capacitors on the board.
 

Offline semir-tTopic starter

  • Regular Contributor
  • *
  • Posts: 56
  • Country: ba
Re: TMC5160 keeps dying on 24V input
« Reply #9 on: May 29, 2023, 08:54:36 am »
I am using the Cheap Chinese power supply. As this power supply doesn't have on/off button I connect it via caples to the power supply when I want to power it on.

So this could be an issue ? How can I avoid this ?

 

Online wraper

  • Supporter
  • ****
  • Posts: 16793
  • Country: lv
Re: TMC5160 keeps dying on 24V input
« Reply #10 on: May 29, 2023, 09:03:00 am »
Without any bulk capacitance on 24V close to the power MOSFETs, voltage can easily overshoot during motor operation.
 

Offline semir-tTopic starter

  • Regular Contributor
  • *
  • Posts: 56
  • Country: ba
Re: TMC5160 keeps dying on 24V input
« Reply #11 on: May 29, 2023, 09:06:33 am »
One thing to note is that the motor is not running when I power the 24V supply. It is idleing.

But I can add large capacitor for sure on the board and test.
 

Offline PCB.Wiz

  • Super Contributor
  • ***
  • Posts: 1473
  • Country: au
Re: TMC5160 keeps dying on 24V input
« Reply #12 on: May 29, 2023, 09:37:54 am »
I am using the Cheap Chinese power supply. As this power supply doesn't have on/off button I connect it via caples to the power supply when I want to power it on.

So this could be an issue ? How can I avoid this ?

It is common for designs to use a separate resistor feed to the driver IC's, as they can tolerate larger series R than the power FETs.
Plus local to IC decoupling C's and even a clamp zener.

For your tests you could try a series R in the cable, with a resistor reduction switch for motor-on testing.
I would also manually ramp the power slowly up and down to/from 24V to check that is ok.
 

Offline semir-tTopic starter

  • Regular Contributor
  • *
  • Posts: 56
  • Country: ba
Re: TMC5160 keeps dying on 24V input
« Reply #13 on: May 29, 2023, 09:43:39 am »
Thanks.

What I will try to do is:
1. Add bulk capacitor to power rail.
2. Add series resistor.
3. Increase the voltage gradually

I have decoupling C near the IC so I should be good on that point.
 

Offline semir-tTopic starter

  • Regular Contributor
  • *
  • Posts: 56
  • Country: ba
Re: TMC5160 keeps dying on 24V input
« Reply #14 on: June 02, 2023, 11:43:40 am »
Just to confirm the solution for someone to find in the future if needed.

I confirmed that the issues was lack of the large capacitor on the power line. I soldered 150uF electrolytic capacitor and gradually increased the voltage. It worked good this time. Then I decided to try hot-plugging it immediately to 24V and it didn't die.

There is a spark when connecting the banana connector to the power supply but I think I should be able to solve this by adding inrush current limitation circuit.
 
The following users thanked this post: thm_w

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 6459
  • Country: nl
Re: TMC5160 keeps dying on 24V input
« Reply #15 on: June 02, 2023, 03:13:35 pm »
So the cause was powersupply overshoot ?
How much and how long does it overshoot , that ic can withstand 60V.
I would throw such a PSU in the trash before it kills more circuits in the future.
 

Offline langwadt

  • Super Contributor
  • ***
  • Posts: 4391
  • Country: dk
Re: TMC5160 keeps dying on 24V input
« Reply #16 on: June 02, 2023, 06:11:56 pm »
So the cause was powersupply overshoot ?
How much and how long does it overshoot , that ic can withstand 60V.
I would throw such a PSU in the trash before it kills more circuits in the future.

probably not the fault of the powersupply.  To maintain the set current in the stepper the driver will both sink and source current from the supply, without enough bulk capacitance that can cause voltage spikes
 

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 6459
  • Country: nl
Re: TMC5160 keeps dying on 24V input
« Reply #17 on: June 02, 2023, 08:38:29 pm »
Do you mean back-emf ?
 

Offline mikerj

  • Super Contributor
  • ***
  • Posts: 3233
  • Country: gb
Re: TMC5160 keeps dying on 24V input
« Reply #18 on: June 02, 2023, 09:33:34 pm »
So the cause was powersupply overshoot ?
How much and how long does it overshoot , that ic can withstand 60V.
I would throw such a PSU in the trash before it kills more circuits in the future.

It's not the fault of the PSU, it's caused by connecting power to a circuit via some length of wire when the circuit has only ceramic capacitors decoupling it's supply rail.  This app note gives the details of the issue: https://www.analog.com/media/en/technical-documentation/application-notes/an88f.pdf
 
The following users thanked this post: thm_w, Kjelt


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf