Author Topic: Auto-Standby circuit doesnt work as expected  (Read 6865 times)

0 Members and 3 Guests are viewing this topic.

Offline MichaelaudioTopic starter

  • Regular Contributor
  • *
  • Posts: 212
  • Country: de
Auto-Standby circuit doesnt work as expected
« on: August 02, 2020, 02:11:00 pm »
So this is an Auto - stby circuit for a 2.1 Ch Amp with 3x TPA3116. I copied this circuit from a H/K Sub so it works (yea somehow) So I assembled the pcb (without the TPAs) and tested the circuit, everything fine the Mosfet is switching off after ~15 mins. But now everything is assembled it doesnt switch off anymore.. the voltage at pin6 on IC9b is 7.7V @24V input and on Pin 5 it rises. So when the voltage at p5 is >7.7 it stays for like 0.1sec over 7.7 and gets back to like 7.6 and it starts over and over and i dont know what it is, the input Voltage is stable and im running out of ideas, HELP!  (Audio signal would Come in on the 2 47k resistors, the voltage on the output at Pin 7 ic9 would stay low if signal applied)
« Last Edit: August 02, 2020, 09:42:25 pm by Michaelaudio »
 

Offline OM222O

  • Frequent Contributor
  • **
  • Posts: 768
  • Country: gb
Re: Auto-Standby circuit doesnt work as expected
« Reply #1 on: August 02, 2020, 02:13:35 pm »
maybe start by posting a schematic of the actual circuit?
 

Offline MichaelaudioTopic starter

  • Regular Contributor
  • *
  • Posts: 212
  • Country: de
Re: Auto-Standby circuit doesnt work as expected
« Reply #2 on: August 02, 2020, 02:16:27 pm »
I was just adding it
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: Auto-Standby circuit doesnt work as expected
« Reply #3 on: August 02, 2020, 09:24:19 pm »
Post the whole schematic, not just part of it — at least not without identifying the inputs, outputs, etc.

(Not sure why you’d even bother posting a screenshot of a part of the schematic here, when in another thread you posted the entire schematic as a PDF!  :palm: )
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: Auto-Standby circuit doesnt work as expected
« Reply #4 on: August 02, 2020, 09:26:15 pm »
You could also just use the TPA3126, which is pin compatible but has much lower quiescent power consumption, then dispense with the whole auto-power circuit.
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: Auto-Standby circuit doesnt work as expected
« Reply #5 on: August 02, 2020, 09:30:30 pm »
Also, on the full schematic, I see that you’ve got the shutdown pins permanently pulled high to VCC, and then you use the auto-standby circuit to disconnect VCC via a mosfet. But you should be sequencing power to mute and shutdown before/after VCC, so as to prevent popping. If you’re not actually using the auto-standby to disable the power supply (which you aren’t), it’d be smarter to leave VCC connected and instead control the shutdown pin. You’ll get pretty much the same power savings, but without the popping. Even without sequencing the mute pin it’d still be superior to just turning off VCC.
 

Offline MichaelaudioTopic starter

  • Regular Contributor
  • *
  • Posts: 212
  • Country: de
Re: Auto-Standby circuit doesnt work as expected
« Reply #6 on: August 02, 2020, 09:40:02 pm »
I wanted to keep the focus on this section of the schematic thats why :)
Oh I have no turn off pop, thats the op amp for on the mute Pin,  and I wanted to stop the op's running too so thats why I did it that way. But yea it wont turn off as I wrote, the voltage on Pin 5 is just ~0.1sec high er than on Pin 6. Still have no clue why 
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: Auto-Standby circuit doesnt work as expected
« Reply #7 on: August 02, 2020, 09:45:58 pm »
I figured that’s why you posted a section, but it’s missing critical information that make it more difficult to analyze, as one has to guess about key things like the signal inputs... :palm:
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: Auto-Standby circuit doesnt work as expected
« Reply #8 on: August 02, 2020, 09:57:11 pm »
FYI, I don’t know if it has anything to do with the auto standby problem, but I see you’ve got the mute pins connected to one of the TPA’s GVCC pins via a big 10μF capacitor — the same GVCC pin the data sheet expressly says to not connect ANYTHING to except for the 1μF bypass cap to ground, and the resistor dividers for configuration. (Datasheet page 4: “Not to be used as a supply or connected to any component other than a 1 μF X7R ceramic decoupling capacitor and the PLIMIT and GAIN/SLV resistor dividers.”)
 

Offline MichaelaudioTopic starter

  • Regular Contributor
  • *
  • Posts: 212
  • Country: de
Re: Auto-Standby circuit doesnt work as expected
« Reply #9 on: August 03, 2020, 06:11:24 am »
It has nothing to Do with my problem, its a well known anti turn on pop methode for the 3116, there isnt much current flow so its ok but yea, I could connect the cap to VCC as well. U doesnt seem to have a clue why auto stby dont work either right?
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 13218
Re: Auto-Standby circuit doesnt work as expected
« Reply #10 on: August 03, 2020, 07:09:00 am »
The timing network appears to be R71 2Meg charging C78 220uF. Its time constant is approximately 7 minutes.  Its decreased a bit by the effective load of Q3 base, which can be approximated as a 30Meg resistor to ground (emitter_load*hFE) so it will only charge to 94% of the supply, less approx 0.7V  Additionally you are comparing it with half the supply voltage so it *should* only delay for approx 5 3/4 minutes. 

The fact you observed a 15 minute delay means either the leakage current of C78 is a large proportion of its charging current, or the reset transistor Q2 is leaky or is getting partially turned on intermittently.

The whole approach of an extremely large time constant RC network with an electrolytic cap is irretrievably flawed due to the unstable and temperature dependant leakage current.  IMHO you need to redesign it with a digital timer, either an oscillator and divider chain, or a MCU.

Consider using a CD4060 configured for a low frequency RC oscillator, with a diode from the desired output to the RS input to inhibit its oscillator at the end of the delay. Drive its MR (master reset) with your activity signal.
 

Offline MichaelaudioTopic starter

  • Regular Contributor
  • *
  • Posts: 212
  • Country: de
Re: Auto-Standby circuit doesnt work as expected
« Reply #11 on: August 03, 2020, 08:18:56 am »
But the circuit works?It just doesnt work when I soldered the ICs on the PCB cause on Pin 5 the voltage drobs as soon as its higher than on Pin 6 on IC9b. I know they r smarter ways to do it but I want to solve the problem not re design it. So the time delay thing is strange yea, but anyway it works (almost)

What I found out so far:
-when the voltage goes up and down (7.7-7.6V at Pin5 IC9B (measured to virtual ground) the voltage on the  Base of Q3 doesnt change, so that means c78 is full charged, so it has something to do with the comparator itself?

-when I remove R78 it works
-when I remove D16/17 it works
Im sure im close but still cant see where the problem is
« Last Edit: August 03, 2020, 08:24:43 am by Michaelaudio »
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 13218
Re: Auto-Standby circuit doesnt work as expected
« Reply #12 on: August 03, 2020, 08:34:34 am »
What do you expect to happen when IC9B acting as a comparator causes the MOSFET Q4 to switch off the Vdd supply to the timing circuit  feeding its own inputs?   The circuit is fundamentally flawed and if it operates at all its by pure luck.
 

Offline MichaelaudioTopic starter

  • Regular Contributor
  • *
  • Posts: 212
  • Country: de
Re: Auto-Standby circuit doesnt work as expected
« Reply #13 on: August 03, 2020, 09:14:53 am »
Oh my fault! I forgot to switch the Power Symbol, the circuit is always powered ofc.
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: Auto-Standby circuit doesnt work as expected
« Reply #14 on: August 03, 2020, 09:27:27 am »
VDD is the unswitched always-on power, and VCC is the switched power, correct?

(And no, I don’t know why it’s not working, sorry! But I’d listen to Ian.M.)
 

Offline MichaelaudioTopic starter

  • Regular Contributor
  • *
  • Posts: 212
  • Country: de
Re: Auto-Standby circuit doesnt work as expected
« Reply #15 on: August 03, 2020, 09:39:28 am »
Correct. And i did put the Symbol on the drain of Q4 so I dont know what Ian.M means, the Timing circuit is always powered ofc.
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 13218
Re: Auto-Standby circuit doesnt work as expected
« Reply #16 on: August 03, 2020, 10:03:11 am »
Oh my fault! I forgot to switch the Power Symbol, the circuit is always powered ofc.
VDD is the unswitched always-on power, and VCC is the switched power, correct?

(And no, I don’t know why it’s not working, sorry! But I’d listen to Ian.M.)
Correct. And i did put the Symbol on the drain of Q4 so I dont know what Ian.M means, the Timing circuit is always powered ofc.
That's  not what your schematic shows.  If VDD was always on, the IFR5505 P-MOSFET would be ineffective for switching VCC as its body diode would conduct when there was no gate drive.

Ergo you haven't built it as drawn, don't understand it and our attempts to help you are as futile as pissing up a rope.

As a reminder and to prevent O.P. being tempted towards revisionism, here's the relevant section of the schematic (annotated with my understanding of its connections).  The only other points worthy of note is that the rail splitter providing GNDA is also fed from VDD.
« Last Edit: August 03, 2020, 10:16:35 am by Ian.M »
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 13157
  • Country: ch
Re: Auto-Standby circuit doesnt work as expected
« Reply #17 on: August 03, 2020, 10:08:10 am »
As I gather from the full schematic from his other thread, Con7 is where the 24V is fed into the circuit, so it’s definitely not switched. But yes, you’re right, the way that mosfet is used, it won’t ever turn off VCC!
 

Offline MichaelaudioTopic starter

  • Regular Contributor
  • *
  • Posts: 212
  • Country: de
Re: Auto-Standby circuit doesnt work as expected
« Reply #18 on: August 03, 2020, 10:55:28 am »
Okay yep! Q4 is wrong on the schematic.. S and D need to be switched, so on the pcb the Mosfet is connected right (S to power supply D to the Diodes)  guess I should update it and post the right schematic.
 

Offline MichaelaudioTopic starter

  • Regular Contributor
  • *
  • Posts: 212
  • Country: de
Re: Auto-Standby circuit doesnt work as expected
« Reply #19 on: August 03, 2020, 07:13:07 pm »
Okay so I figured out, that the voltage drop after the voltage goes above 7.7V on Pin 5 is the Voltage of the hysteresis. If I replace R76 with a 100K the voltage drop is 1.2V with 1M its 0.1 so its obviously the hysteresis voltage.
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 13218
Re: Auto-Standby circuit doesnt work as expected
« Reply #20 on: August 03, 2020, 07:28:32 pm »
What's the state of the pin 7 output after you observe this voltage drop?  If the OPAMP is correctly wired as a comparator with hysteresis, you should get a step change of the pin 5 voltage upwards, as the rising input voltage crosses the reference level on pin 6.
 

Offline MichaelaudioTopic starter

  • Regular Contributor
  • *
  • Posts: 212
  • Country: de
Re: Auto-Standby circuit doesnt work as expected
« Reply #21 on: August 03, 2020, 07:40:25 pm »
So the Voltage at Pin 6 is 7.7V when the voltage at Pin5 is at lets say 7.71 Pin 7 is at ~23V (the LED blinks for like half a sec) after that the output is ~0.7V and the voltage on Pin 5 is 7.6V. When i take R78 out (Gate resistor) The voltage rises just fine and the output is constant. I tried a higher value for R78 but changed nothing. Really weird..
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 13218
Re: Auto-Standby circuit doesnt work as expected
« Reply #22 on: August 03, 2020, 08:04:09 pm »
What happens if you disconnect R69 (base drive to the timing reset transistor Q2)?

Where is IC9 getting its supply from?
 

Offline MichaelaudioTopic starter

  • Regular Contributor
  • *
  • Posts: 212
  • Country: de
Re: Auto-Standby circuit doesnt work as expected
« Reply #23 on: August 03, 2020, 08:24:57 pm »
Havent tried that yet.
Thats getting the Power from VDD (directly from power supply)
 

Offline MichaelaudioTopic starter

  • Regular Contributor
  • *
  • Posts: 212
  • Country: de
Re: Auto-Standby circuit doesnt work as expected
« Reply #24 on: August 04, 2020, 04:41:46 pm »
Okay verry interesting things going on here. When I disconnect R69 while circuit powered up, Q4 opens! When I then connect it again, Q4 is closing (low on Pin 7 IC9b). When I turn psu off, wait 10 seconds (resistor disconnected) turn it on again Q4 still open. Verry interesting
Btw I replaced R71 with a 100K to Charge te cap faster, shouldnt be a matter
« Last Edit: August 04, 2020, 05:23:40 pm by Michaelaudio »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf