LM2596 gets extremely hot

More
3 months 1 week ago - 3 months 1 week ago #11420 by Joker
Replied by Joker on topic LM2596 gets extremely hot
with the stk500, i can connect to the atmel and read the memory. The STK500 didn't went crazy. I think the Atmel in not the problem.

i have checked all resistors and diodes for orientation and value.

mybe one of the coils is not the best one.
L1: www.conrad.de/de/p/bourns-rlb0912-101kl-...-a-1-st-1056023.html

L2: i can't find the ordered par but the mark says 330 this should be 33µH

the resistors R9 and R10 are not mixed up
I just noticed that the capacitors c5 and c6 have the inscription 221, this should correspond to 220 pF, according to the wiring diagram 22 pF is required (220). But I wonder if this could be the reason if the Atmel does not get a clean reference clock.
Last edit: 3 months 1 week ago by Joker.

Please Log in or Create an account to join the conversation.

More
3 months 1 week ago #11421 by Ty_Eeberfest
The coils seem okay to me, as long as you are sure you have the 33uH in the low voltage circuit and the 100uH in the high voltage circuit.

Those 220pF / 22pF capacitors can definitely make a difference in the quality of the crystal cpu clock. I have never experimented to see how much of a difference using the wrong caps makes but I strongly suggest you put in 22pF. This may or may not make a difference to your power problem, but when any part is 10 times the specified value it is suspect! In the past we have heard of strange random behavior caused by problems with the crystal and its loading caps.

Look into it later when the dust is clearing off the crater.
The following user(s) said Thank You: Ian

Please Log in or Create an account to join the conversation.

More
3 months 1 week ago #11423 by Ian
Replied by Ian on topic LM2596 gets extremely hot
Oh yeah, that's a really good point Ty.

The crystal oscillator can be a huge source of trouble - I have seen two possible sources of trouble:
  1. Defective/out of spec load caps - This one is what Ty is referring to - it can cause the oscillator frequency to be unreliable
  2. Noise - I have observed this in a couple of cases: the circuit freaks out for no apparent reason

Take a wire from the GND line of the circuit and ground the can of the crystal with it. You can measure the high voltage with a meter to see if it is stable.

If this doesn't help, then we should take a look at those load caps.

Please Log in or Create an account to join the conversation.

More
3 months 1 week ago #11424 by Joker
Replied by Joker on topic LM2596 gets extremely hot
I just checked my order. Actually, it should be 22nF.

I had this once with a clock (internal load resistor of the Atmel activated and additionally external capacitors attached). As a result the clock rans too slow.

With factor 10 I would have suspected that the Atmel runs very slow or the crystal doesn't oscillate at all. I wouldn't have expected the power supply to tick out.

I will definitely replace the capacitors. Unfortunately my electronics shop is closed because of Corona regulations. I'll have a look in the grab box on work on monday.

Please Log in or Create an account to join the conversation.

More
3 months 1 week ago - 3 months 1 week ago #11426 by Ian
Replied by Ian on topic LM2596 gets extremely hot
OK, please come back with the result. 22nF is definitely way out of spec. I'm amazed it runs at all...

Thanks!
Last edit: 3 months 1 week ago by Ian.

Please Log in or Create an account to join the conversation.

More
3 months 1 week ago #11442 by Joker
Replied by Joker on topic LM2596 gets extremely hot
The capacitors are exchanged, unfortunately the board still behaves the same.

The error must be somewhere in the HV circuit. The HV starts at 40V and rises slowly to about 150V. Then the current consumption is 500mA and it starts to make noise. Here I stopped, I set the current limitation to 500mA. I have as said no load on the board.

C5 22pF replaced by 22pF ceramic
C6 22pF replaced by 22pF ceramic
Q2 16MHz imprint fits
C7 100nF imprint 104 - fits
C8 100nF imprint 104 - fits
D2 UF4007 tested - direction fits
S28 SOCKET 28 without words
C1 2.2uF 400V tested - direction fits
L1 100uH soldered out, imprint 101 - fits also
R7 3k measured again - psst
Q1 IRF840 imprint fits
R9 390k measured - fits
R10 4.7k measured - fits
R18 10k remeasured - fits
IC2 MEGA328-P imprint fits

i have no idea. Either C1 is defective or it's because of the atmel. I have compiled the program in d Arduino ide and transferred the hex file using (STK500 and ISP). Can I test if the Atmel works correctly? Can I check something via the Wlan module?

Please Log in or Create an account to join the conversation.

Moderators: AccutronTy_EeberfestIan
Time to create page: 0.094 seconds

Search

Tube Suppliers

Go to top
JSN Boot template designed by JoomlaShine.com