SG3525 cant make it work

[DarkZero]

Hey All!

i made this circuit http://electrodb.ro/wp-content/uploads/2011/04/112-500x383.jpg i checked everything like 10x am i missing something ?

[Sirius631]

Pin 10 should be tied high, not low. The datasheet has the shutdown active low, so you are keeping it shutdown, so it won't work.

[hamdi.tn]

On the contrary circuit shut down when connected the any voltage upper than 0.8V

[DarkZero]

Pin 10 is shutdown. When this pin is low, PWM is enabled. When this pin is high, the PWM latch is immediately set. This provides the fastest turn-off signal to the outputs. At the same time the soft-start capacitor is discharged with a 150µA current source. An alternative method of shutting down the SG3525 is to pull either pin 8 or pin 9 low. However, this is not as quick as using the shutdown pin. So, when quick shutdown is required, a high signal must be applied to pin 10. This pin should not be left floating as it could pick up noise and cause problems. So, this pin is usually held low with a pull-down resistor.

every tutorial on this chip says this and i tried making it high

[mrkev]

Pin 10 should be tied high, not low. The datasheet has the shutdown active low, so you are keeping it shutdown, so it won't work.

I wouldn't say that is the problem. In the first picture of the datasheet, shutdown is shown as active when low. But when you actually scroll down, you'll find out that shutdown is active when HI... Viz :
An alternate approach is the use of the shutdown circuitry of Pin 10 which has been improved to enhance the available shutdown options. Activating this circuit by applying a positive signal on Pin 10 performs two functions: the PWM latch is immediately set providing the fastest turn–off signal to the outputs; and a 150? A current sink begins to discharge
the external soft–start capacitor.

The problem is IMHO at R7 & R6, try to disconnect both.

[hamdi.tn]

+1 for R6 and R7

[mrkev]

Yeah, I've look at it once more and I am almost certain...
Whole purpose of pins 1, 2 and 9 is to have a high impedance voltage sense for the comparator. It should be used as voltage follower (pins 1 and 9 tied together, with optional resistor).

The way you've got it, it's kind of really weird inverting amplifier. So since there is about 2,14V at pin 2, there is gonna be (almost) 0V at the comp. input and therefore it's set for 0 duty cycle.

Keep in mind that voltage at pin 9 corresponds to the duty cycle:
PWM COMPARATOR SECTION (in datasheet)
Input Threshold, Zero Duty Cycle 0.9V
Input Threshold, Maximum Duty Cycle 3.3V

I am not sure if you can override the OA (and set voltage at pin 9 directly), but when you connect OA it as follower, the voltage at pin 2 will be one to set duty cycle.

[Sirius631]

Quote from: Sirius631 on Today at 07:06:56 AM

Pin 10 should be tied high, not low. The datasheet has the shutdown active low, so you are keeping it shutdown, so it won't work.

I stand corrected over this. I simply looked at the first pinout on the data sheet and saw that it had a bar over 'Shutdown'. Convention was that this indicates an active low control. Reading further into the text says the exact opposite.

http://www.farnell.com/datasheets/19849.pdf

[DarkZero]

Hey all!

good and bad news, i did remove those resistors and i got wave forms like in attached pictures one of the channels seems to work the other is not, faulty chip? or am i still missing something.

though i noticed  those spikes they seem to come from my PSU (Computer PSU) when i add 220uF on the 12v supply rail PSU changes sound and that channel stops working, im starting to suspect bad PSU.

[mrkev]

Hey all!

good and bad news, i did remove those resistors and i got wave forms like in attached pictures one of the channels seems to work the other is not, faulty chip? or am i still missing something.

though i noticed  those spikes they seem to come from my PSU (Computer PSU) when i add 220uF on the 12v supply rail PSU changes sound and that channel stops working, I'm starting to suspect bad PSU.

Hi again. I didn't really get much from that post, so hard to guess
You are powering whole thing from PSU? If yes, the problem may be that this is too tiny load, so I would try to add something like >2W 100ohm power resistor to the 12V rails. You could measure the PSU output directly with your scope, to see what's happening there. That supply could be completely OK... And since you are drawing only like few mA, you can try changing R10 to about 47 - 100 ohm.

The second channel not working is weird. It could be caused by so many different things (short on board, mistake in measurement method, damaged output /f.e. if it was shorted before/, cold joint, wrong connection of probes, schematic inconsistent with board, wrong output resistor value etc.), but not by the mistake in the schematic.
They are internally connected and if channel A works, channel B should work too. Even if you have bad PSU, if you get correct waveform from one channel, you should get it from the second one too. What's more, it's highly improbable that you did get chip that was faulty from the supplier. That is something that almost never happens, especially with not ESD sensitive chip like that.

So I would look for mechanical stuff (like copper "dust" between leads, cold joint), checked the connection and value of R8&R9 (you didn't wrote which channel is not working) and probed directly on the SG3525's pin. And you could also try to power it from another supply (9V battery could be fine for short time, check the input current). PSUs are great powerful sources, but they can have problem with earth loops, especially if you are using scope that is earthed by mains.

[DarkZero]

Hi Mrkev!

Guess i need to provide more info

Picture one shows output B, Picture 2 shows output A i measured them directly on the pins which arent connected to anything.
Also i get same waverform as in picture 2 when i probe supply rail except its at 12v , both scope and PSU are not earthed. PSU is of the older type maybe it does need that minimum load to work properly i will give that a try. i will try things you suggested because they make sense hope i figure it out.

[DarkZero]

Hey all!

It works, it was bad contacts. i attached schematic i finally used

Thanks all !

[mrkev]

Glad I could help Btw. if you wanna vary the duty cycle, use +- 10k pot instead of R9.

[hamdi.tn]

Great  can i ask for what purpose are you using it ?

[DarkZero]

hey all!

at the moment i don't need to vary duty cycle but ill keep it in mind for some other project i mite need it !

im trying to build 300W car amplifier and i need some 75V at the output that's why i'm making this oscillator to step up voltage, transformer is problem i used core from high voltage transformer found inside old tv's 15-15 turns in primary and i wound secondary until i got 90V-0-90V but voltage drops to 15V with 100W load and of course i didn't drive transformer from this circuit i have irfz48 x 8 connected to this circuit. all tutorials on transformers require some data that manufacturer gives so i'm not really sure how to over come this ^^

[hamdi.tn]

i recently did a project of 200W UPS with SG3525 ( due to manufacturing i still didn't receive the prototype to test it  ) , i used a calculator to help me design the step up transformer since i have zero experience on how to design one, i did some maths too to confirm the result of the calculator, i share the little calculator , you may find it helpful 

[DarkZero]

haha this program... i bet its nice but

[BennVenn]

Check that your core isn't gapped, that is, when both halves are together there is zero gap between them. It is likely the core you salvaged is from a flyback converter which won't work all that well in a push-pull design.