Halfbridge with an inductive load of 24 µH

Thread Starter

Alex009

Joined Jul 2, 2014
33
Good morning everyone,

I could Need a Little help with the attached circuit.

So the basic idea is that the halfbridge driver drives the 2 nmos-gates. So there should be always 1 active (well atleast if you dont consider the fact that there is a deadtime). This should be done at a frequenzy of 125 kHz (T ~ 8 µs). So I got a variable R2 of 5.5 k Ohm (potentiometer) and the capacitor C3 to make sure the halfbridge driver works at this frequenzy.

--> I thought my 24 V DC would Change to a 0V to 24 V square wave (right before C2) and then Change to an -12V - 12 V AC (right before L1). But unfortunately when getting the whole thing done on a testing board nothing happend. I simple did not get a switching Signal out of the halfbridge driver ( --> neither did i get a square wave at one of the mosfets).

So do you have any ideas or suggestions to help me with that issue? I will add the mosfet as soon as I m able to find out which ones I used (I'm currently not in reach of these).
Maybe changing some values of c's and r's?


I m using the following halfbridge driver: IRS2153 http://www.irf.com/product-info/datasheets/data/irs2153d.pdf

Best regards
Alex
 

Attachments

Last edited:

ScottWang

Joined Aug 23, 2012
7,397
When M1 is turn on then Vds=0V, Vd=Vs=24V.
The condition to keep M1 turn on that it is Vgs = 5V or 10V, so the Vg to ground will be Vg = 24V+5V=29V or Vg=24V+10V=34V.
 

Alec_t

Joined Sep 17, 2013
14,280
There could be various reasons why your breadboard circuit didn't work, but the LTspice simulation won't work properly unless you choose a value for the NMOS FETs.
 

Thread Starter

Alex009

Joined Jul 2, 2014
33
@ScottWang well thats exactly my Problem. gate-source voltage is always ~0 V for both Mosfets
@Alec_t Yes it doesnt. There is also no LTSpice model for the halfbridge driver - I did this in LTSpice for you to understand the circuit i m using (The IRS2153 is ONLY REPRESENTED by a standard 8 pin package) - since i didnt get a real one on the Internet. It would be helpful if you could give me a hint concerning on of these various reasons.
 

Alec_t

Joined Sep 17, 2013
14,280
Breadboarded circuit failure, possible reasons:
Incorrect/missing connections, high resistance connections, faulty/fried component, wrong component value, stray capacitance, stray wiring inductance, .....
 
Last edited:

Thread Starter

Alex009

Joined Jul 2, 2014
33
Breadboarded circuit failure, possible reasons:
Incorrect/missing connections, high resistivity connections, faulty/fried component, wrong component value, stray capacitance, stray wiring inductance, .....
Well since i double checked Connections and was measuring several points i dont see this to be the case. Also I was double checking the components i used. They all fit the schematic. So thats why I asked if there is anything wrong with it.
 

Alec_t

Joined Sep 17, 2013
14,280
You say you are using a IRS2154, but the link is for a IRS2153 ?? Is the pin-out and function the same?
What waveforms/voltages do you get at pins 3, 5 and 7?
 

Thread Starter

Alex009

Joined Jul 2, 2014
33
Oh ty for telling me. The link is actually the correct one. Just a typo. I will edit this asap.

I got on both 5 and 7 a ~200 mV Voltage which looks like noise to me.
 

Alec_t

Joined Sep 17, 2013
14,280
I got on both 5 and 7 a ~200 mV Voltage
Then it looks like there is a wiring error, a supply voltage/ground missing, or a dead/fake IC. You could try re-positioning wiring on the breadboard, in case there is a dodgy socket.
What was the pin 3 waveform like?​
 

Thread Starter

Alex009

Joined Jul 2, 2014
33
Hey I will Report back on monday and upload the Pictures of the waveforms since I m not having acces to a scope right now. Everything is closed here til monday :(
I will also try to use another IC since i ordered a few of them.
Have a nice Weekend and ty so far.
 

Alec_t

Joined Sep 17, 2013
14,280
From the IC datasheet, the recommended max supply current Icc is 5mA and the minimum Vcc clamp voltage is 14.4V. With R3=470Ω and V1=24V your Icc could be (24-14.4)/470 = 20.4 mA ! That is only just over the absolute maximum allowable, but may well have fried the IC :(. I draw your attention to Note 1 at the foot of page 2 of the datasheet. Perhaps make R3=2k2 to 4k7?
 
Last edited:

Alec_t

Joined Sep 17, 2013
14,280
I couldn't find a model of the IRS2153 anywhere on the web, so I built one. I've attached it below, in case it helps. It has a few limitations (described in the attached .sub file, readable with a text editor).
 

Attachments

Thread Starter

Alex009

Joined Jul 2, 2014
33
@Alec_t Thank you very much for your help. As promised i will provide you with the waveforms of all out-/Input Pins of the IC.
For doing that I also simplified the circuit as shown in the Picture below.


Pin 1:



Pin 2:


Pin 3:


Pin 5:


Pin 7:


Pin 8:


All waves except for the one on pin 7 are looking good to me. I also tried to Change the capacitor c5 to a slightly higher value (used 1µF instead) but this didn't change anything. I figured this was also the Problem my circuit suffered when using the mosfets.
The frequency is also alright. I initially wanted a frequenzy of ~125kHz so I will just change the valuce of c3. I also do not think that this is the Major Problem with this. Since when changing the capacitor to a much higher value (operating frequency ~5kHz) the pin 7 "Problem" still occured.

Would be nice to hear from you again.

BR
Alex

Edit: Also thank you for the hint concerning R3. Will probably chose 1k Ohm.
 
Last edited:

Alec_t

Joined Sep 17, 2013
14,280
It's not surprising pin 7 doesn't do the same sort of switching as pin 5. Pin 6 should not be grounded. The bootstrap can't work unless the lower end of C5 can be pulled high and low.
With pins 5 and 7 each grounded via 33 Ohms they will have an excessive load, since they are rated to source only ~180mA.
 
Last edited:

Thread Starter

Alex009

Joined Jul 2, 2014
33
@Alec_t So what would you suggest for connecting Pin 6 to for testing? But obviously the IC works perfectly fine . My next guess was changing my Cboot to a higher value ~5µ. What is your thought on that?
Thank you for your help.
BR Alex

Edit: Could the capacitive / inductive load cause such Problems?
Could the high frequency cause Problems?


I' m using the following Mosfet: STP36NF06 - http://www.farnell.com/datasheets/1718576.pdf -. What are your thoughts on this one?
 
Last edited:

Alec_t

Joined Sep 17, 2013
14,280
So what would you suggest for connecting Pin 6 to for testing?
Connect it as in post #1.
My next guess was changing my Cboot to a higher value ~5µ. What is your thought on that?
I'd leave it at 100n, unless you find the boot voltage at pin 8 drooping.
But obviously the IC works perfectly fine
So what problem are you referring to?
That FET looks fine.
 

Thread Starter

Alex009

Joined Jul 2, 2014
33
@Alec_t Yes. I think I found the Problem and going to solve it hopefully 'til tuesday. I accidently bought the wrong capacitor for CT and didn't realize somehow.. It should have had ~1nF instead I bought 0.1 nF and used it on my breadboarded circuit. I'm still wondering why I even got some o.k looking signals :'D Lets hope the whole thing will work with the new capacitor on tuesday.
Oh Boy.. Thank you anyways. I will keep you informed.

BR Alex
 

zealthy

Joined Jun 27, 2019
1
Hi Alex,

I know it's already 4 years ago, but I encounter the exactly same problem, the high side signal on pin7 is missing, did you solve this problem eventually?

BR Julian
 

kubeek

Joined Sep 20, 2005
5,794
Not sure what your issue is, but the Alex009 is missing the bootstrap diode to charge C5 and he most likely didn´t have the D version of the chip.
 
Top