LTspice simulation of H-bridge using IR2110

ScottWang

Joined Aug 23, 2012
6,988
When I doing the simulation have something problems, so I cut off the mosfets to do the testing, and I found the the waveform of LO is correct as LIN, but the waveform of HO was different with HIN, the waveform of HO shows as below.

IR2110Hout_circuit.gif

The waveform of HO.
IR2110Hout.gif
 

Attachments

ScottWang

Joined Aug 23, 2012
6,988
The below circuit is working correctly, but when I testing the circuit without mosfets circuit as #23, it shown the HO had some different waveform, so when we using the simulation software always have the blind points, sometimes in the real world can be oscillate, but in the software there is no any oscillation.

@Alec_t :
Do you know how to explain the situation on #23 and #24?

IR2110simNOTworking_Alec_T_circuit.gif
 

Attachments

Alec_t

Joined Sep 17, 2013
11,900
when I testing the circuit without mosfets circuit as #23, it shown the HO had some different waveform
1) Not surprising. Without the MOSFETS switching, the bootstrap circuit won't work correctly.
2) V2 and V3 should be out-of-phase non-overlapping clocks.
3) According to the datasheet the optoisolator outputs should have pull-up resistors when driving CMOS inputs.

Edit:
4) The circuit in post 24 won't work correctly when the default FETs are used, since they draw only ~1mA and result in VS staying at about 12V. This also prevents the bootstrap working. Shouldn't your .inc directive include the IRF540 sub file?
 
Last edited:

ScottWang

Joined Aug 23, 2012
6,988
1) Not surprising. Without the MOSFETS switching, the bootstrap circuit won't work correctly.
2) V2 and V3 should be out-of-phase non-overlapping clocks.
3) According to the datasheet the optoisolator outputs should have pull-up resistors when driving CMOS inputs.

Edit:
4) The circuit in post 24 won't work correctly when the default FETs are used, since they draw only ~1mA and result in VS staying at about 12V. This also prevents the bootstrap working. Shouldn't your .inc directive include the IRF540 sub file?
When I tried to using the IRF540.sub you provided, but it won't works and it can't be reedited, so I changed to used the internal provided by Linear.

Now the problem in #23, whatever I tried to added a inverter or used a bjt as inverter, but the HO still output the same waveform, but I didn't do anything for LIN and its input/output waveform are the same.

So, now I think there is something wrong at the HIN and LO in the IR2110g you provided.

IR2110simNOTworking_Alec_T2-half_circuit.gif
 

Attachments

Alec_t

Joined Sep 17, 2013
11,900
Scott, you need the VS pin switched by the FETs for the bootstrap to work. Unless VB can exceed VS by ~7.5V the under-voltage protection kicks in and prevents HO going high.
Try my .asc in post #18.
 

ScottWang

Joined Aug 23, 2012
6,988
Scott, you need the VS pin switched by the FETs for the bootstrap to work. Unless VB can exceed VS by ~7.5V the under-voltage protection kicks in and prevents HO going high.
Try my .asc in post #18.
Thanks.
I forgot to checked the internal structure, that's why I felt that very strange when I tried the circuit in #24 is ok, but in #23 and #26 all have the same problem.

I connected a 10 Ohms resistor to Vs and ground to tried, now it is ok, I also connected the Vs to Ground to tried, the output is ok.
 

Thread Starter

afwef1

Joined Apr 1, 2016
15
I'm stuck again. Now I need to connect 3 H-bridges in series so that their voltages add up. For example,we did one H-bridge with VCC = 24V, now I need to have 3 so that voltages add up and we have 72V on our resistor. I have been stuck on this for last 3 weeks, but simply can't find solution. I think that problem is that I can't make 3 separate grounds. Any idea how can I do that ( make 3 separate grounds)?
 

ScottWang

Joined Aug 23, 2012
6,988
Please attach the circuit, including *.asc and *.jpg files, compress the jpg file to a clear and properly size and resolution.
 

Thread Starter

afwef1

Joined Apr 1, 2016
15
This is best picture I could do, schematic is bit large.




In the .asc file you can see that there is one more bridge that is connected the same way as these two are. What I need to get is sum of those 3 voltages (24V + 24V + 24V = 72V ), but according to my mentor , problem is that I don't have separate grounds for every H-bridge, which really is true, LTSpice looks connects every ground on same point. However, even with extensive searching/googling, I couldn't find how to solve that problem, even using some other simulation tool.
 

Attachments

ScottWang

Joined Aug 23, 2012
6,988
Because this is a homework, so I move to homework forum, and in this forum, the members can't give you the dirrect answer, only could give the hint and guide you to the target.

How to solve the problem of ground isolation, you should treat the total power as assembly from the batteries then what will you do for the circuit of each part.
 

Thread Starter

afwef1

Joined Apr 1, 2016
15
How to save the problem of ground isolation, you should treat the total power as assembly from the batteries then what will you do for the circuit of each part.

Sorry I don' think I understand this ? Could you maybe explain it with bit more details?
 

ScottWang

Joined Aug 23, 2012
6,988
Sorry I don' think I understand this ? Could you maybe explain it with bit more details?
I mean that you have to prepare three sets of power, each sets of power providing the isolated power for each h-bridge circuit and its related circuit, otherwise you can't get three isolated ground.
 

Alec_t

Joined Sep 17, 2013
11,900
For stacking the three voltages you may need isolated power supplies. For simulation, the COM label can be useful for nodes where there are commoned connections which aren't at true ground.
COM-label.PNG
 

Thread Starter

afwef1

Joined Apr 1, 2016
15
I have tried using COM label, naming one "ground" COM for one H-bridge. other one COM2 for 2nd H-bridge and leaving third H-bridge on regular ground. When I go to highlight net I can see that they really are separated, but I am still not getting desired ( voltages are not adding up). I am not even sure that this principle can work as he ( mentor ) imagined it.

I included .asc file with COMs as grounds.
 

Attachments

ScottWang

Joined Aug 23, 2012
6,988
It seems there is something wrong with the power of h-bridge, you should be in series from h-bridge 1, h-bridge 2, h-bridge 3, but you in parallel them then the whole h-bridge only can be handle 24V was not 72V, and from the h-bride 1 label ground as gnd1, gnd2 gnd3.
 

Thread Starter

afwef1

Joined Apr 1, 2016
15
It seems there is something wrong with the power of h-bridge, you should be in series from h-bridge 1, h-bridge 2, h-bridge 3, but you in parallel them then the whole h-bridge only can be handle 24V was not 72V, and from the h-bride 1 label ground as gnd1, gnd2 gnd3.

Do you know how it should be connected then ? It looked like parallel connection to me also but mentor said that connection is alright, so I didn't ask any further about it. How should I connect it to makes it series connection ?
 
Top