1ED3323MC12N Problem with "RDY"-Signal

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
lock attach
Attachments are accessible only for community members.
Alex_S
Level 2
Level 2
First solution authored 10 replies posted 10 sign-ins

Dear Community,

I made some new Driver-PCB´s using the 1ED3323MC12N, those are driven (5V) via voltage translation (SN74LVC8T245PW) using a 3.3V-micro-controller. The older PCB using the BM6105 from ROHM, but they stopped production of this this type ...

I attached the schematic of Driver-Circuit. Actually "DESAT"-Function is disabled by an 0Ohm-Resistor instead of the 100pF-Capacitor. Switching frequency (inputs) is 20kHz. Drivers are interlocked for High-Side and Low-Side.

Now to the problem, if I run the system I`m getting Low-Pulses (500ns … 25us) on the "RDY"-Pin of some Drivers, but I don´t know or understand why that occurs.

Referring the datasheet, I tested the following possible sources.

  • the supply voltages on both sides (Vcc1, Vcc2 and Vee2) but no problem there
  • Internal Signal transmission
    1. Fixed ON-Time to mid. value of (uC)          => no more Low-Pulses on “RDY”
    2. Fixed ON-Time to min. value of (uC)          => no more Low-Pulses on “RDY”
    3. Fixed ON-Time to max. value of (uC)          => no more Low-Pulses on “RDY”

 

Do you have an idea the solve this, or some suggestions ?

 

Thanks in advance,

Alex

 

0 Likes
1 Solution

This is very interstig case.

DCDC swithing is noise souse. A 3~10pF between DCDC first and second side. If you see it on the GND, second output is like pingpang jump on the GND.

If your loop is single line, it is high resistance for GND. So a lower capactaticy will be like short with high frequently.

All DCDC second side will been like a "happy jumping tree". Their dance is...

_0-1671452874408.png

 

View solution in original post

0 Likes
15 Replies
喜马拉雅之雪
Employee

Dear Alex: 

Thanks for your comments. I had read your SCH. Would youlike share your waveform?

For DCDC input, looks like that it is very special, would you like share DCDC SCH and your test board picture?

0 Likes
lock attach
Attachments are accessible only for community members.

Good Morning,

 

thanks for your interest, and your reply. In case of DCDC-Converter we´re talking abount an "RECOM - R24P22005D/P", which worked fine with the drivers from ROHM.

The pictures of the board that you asked for, I will take in the evening.

I attached the measured waveforms. But in case of VCC2 and VEE2 the measurement was done against System-GND (left side of schematic), this is the reason for the spikes in the supply voltages.

Thank very much.

 

BR

Alex

0 Likes
喜马拉雅之雪
Employee

Hi Alex:

There are great and beautiful👍

As I'd feared, R24PX DCDC is very professional with standred drive.  For very speedy drive application, as EMI bug, it have some problem. 

_0-1671439580711.png

For the DCDC, the efficient had be set to one of the main parameters, but inside switching had been seting to hard switch and very high dv/dt.

_2-1671440320376.png

 

 

0 Likes

HI,

thanks for your reply.

I took "Note5" of the Datasheet under the pictures you attached. Because Powermodule is an CCS050M12CM2 (SiC).

But as I know the transients will be gone if I´m measuring against GND on High-Voltage-Side insted of Low-Voltage-Side. (I´ll check that in the evening)

In case of very high dv/dt ok, but the measurements were made without an DC-Link-Voltage. But I´ll try to check that in the evening too.

 

Thank you very much.

 

BR

Alex

0 Likes

This is very interstig case.

DCDC swithing is noise souse. A 3~10pF between DCDC first and second side. If you see it on the GND, second output is like pingpang jump on the GND.

If your loop is single line, it is high resistance for GND. So a lower capactaticy will be like short with high frequently.

All DCDC second side will been like a "happy jumping tree". Their dance is...

_0-1671452874408.png

 

0 Likes
lock attach
Attachments are accessible only for community members.

HI,

so I did the measurement with (Scope) GND-Connection on HV-GND. See attachment, do not wonder about that fact that if "RDY" has an falling edge the gate signal would be shut down. (Done by Hardware!)

Sure there are spikes in case of switching, but I´d say if the EMI caused by switching is the problem the falling edge of "RDY" is nearer the switchiing event ?!

As before, thanks for your interest and reply

 

BR

Alex

0 Likes

NOTE: I didt not have an 470pF capacitor, instead of this I used an 220pF ceramic type !

Just a minute ago I tested with capacitor connected as can be ssen in the upper picture, but same behavior as before. Furthermore falling edges on "RDY"-Signal without identifiable source.

0 Likes
喜马拉雅之雪
Employee

For the case, I have 2 idea. Would like try them?

Change loop for EMI

_3-1671443163165.png

 

0 Likes

In the case that nothing will change the situation, I have to... with new PCB´s ...

Thank you

BR

Alex

0 Likes

NOTE:

  • I didt not have an 100pF capacitor, instead of this I used an 220pF ceramic type !
  • Filter (of outputs) impossible to apply afterwards !

Just five minutes ago I tested with capacitor ("CS") connected as can be ssen in the upper picture, but same behavior as before. Furthermore falling edges on "RDY"-Signal without identifiable source.

0 Likes
ShashankK
Employee
Employee
First reply posted Welcome!

Hi Alex,

In order to speed up this solution and to make sure we can better understand your problem, i want our local FAE to contact you. Could you please send me a private message with you email and affiliation, and a local Infineon FAE will reach out to you.

Thanks.

BR,

Shashank

0 Likes

UPDATE:

Meanwhile I did some more measurements and tests (Listed below), but none of that solves the problem. 😩

😭 So the search goes on ....

0 Likes
喜马拉雅之雪
Employee

Hi Alex: Good new year!

would you lke share the IC code or made the IC picture?

0 Likes
lock attach
Attachments are accessible only for community members.

Hi,

merry christmas and a happy new year !

 

Based on the fact I´m not sure of which one I should take a picture, I took a picture from anyone. (All have the same problem).

Additionally I attached some more information in form of a pdf.

 

Thanks for your help!

 

BR

Alex

0 Likes
Alex_S
Level 2
Level 2
First solution authored 10 replies posted 10 sign-ins

Hi,

today the new bypass capacitors (1uF) for driver (C9,10,11 out of "RDY-Fault.pdf" page 2) has been delivered, so I tested them, but nothing changes.

The edges on "RDY" are still there.

 

Thanks in advance

Alex

0 Likes