ir2110 problems

Thread Starter

ian123

Joined Aug 24, 2011
73
please could someonehelp me with this design.it Doen sem to work at all
if i test the Ho is always on on bothsides of the circuit and I cant seem to switch the Lo side on at all.

Iam using a pic12f509 but not in pwm just switching on Q1 andQ3 together waiting 2secs switching off and then switching Q4 and Q2 together. but cant seem to get i going.
 

Attachments

praondevou

Joined Jul 9, 2011
2,942
R1 and R2 should have low resistance like R3 and R4.

Connect Vdd to 5V (logic power supply) and Vcc to a voltage between 10V and 20V. (usually 15V)

The IR2110 has an undervoltage protection for the output voltage supply of about 8V. (see datasheet)
 

SgtWookie

Joined Jul 17, 2007
22,230
IR2110 datasheet: http://www.irf.com/product-info/datasheets/data/ir2110.pdf
You're trying to use 5v for the gate supply drive (Vcc), but the minimum is 10v, max is 20v. You really need at least 11v in order to fully charge your MOSFET gates; as they are not logic level. 12v would be fine. If Vcc is not at least 10v, the IR2110 will shut down to protect your MOSFETs. This is known as an "undervoltage lockout".

You need Vdd to match your logic input voltage. Vdd must not be less than 3.3v. 5v will work fine.

You should really have caps next to the anodes of your boost diodes (D5, D6) so that they get charged quickly; otherwise the inductance of the wiring will delay their charging.
[eta]
Whoops, cross-posted with praondevou - he's right, your R1 and R2 are much too high in value.

I hope you are using dead-time delays between turning off your high-side MOSFETs before turning on your low-side MOSFETs and vice-versa - or you will experience the dreaded "shoot-through" condition, which will quickly destroy your MOSFETs.

Your 12v supply does not appear to be regulated. This will be a problem; if the IR2110 Vcc input is not 10vc to 20v, it will go into lockout mode.
 

Thread Starter

ian123

Joined Aug 24, 2011
73
R1 And R2 actually have the right value 3.3 ohms just made a mistak in the drawing I will make necessary changes and test thank you
 

Thread Starter

ian123

Joined Aug 24, 2011
73
Sgt wookie

Please advis e if the cap should go to ground from the anode and will a 0.1uf do the trick

Regards
Ian
 

SgtWookie

Joined Jul 17, 2007
22,230
Yes, the caps should go to ground from the anodes.

You are showing 10uF for your boost caps. The caps on the anodes of the boost cap diodes will be charging the 10uF caps. How much do you think that you will need?

Placing a 10uF and 1uF in parallel would likely work, depending on how deeply you are discharging the boost caps.

In order to charge the boost caps, you will need to turn your low-side MOSFETs on first for at least a short period of time. Otherwise, there is no ground path for the boost caps, and your high-side MOSFETs will not turn on. This might not be apparent at first, but if you trace out the current paths for the boost caps, you'll see that the only path to ground for the low side of the boost caps is via the low-side MOSFET.
 
Last edited:

Thread Starter

ian123

Joined Aug 24, 2011
73
Ok have tried the caps on the anode and turned on the lo side first for 10millisecs but still no HO , After the first mods the LO side is Switching but the HO side not. I have tried a bigger cap forC1 And C2 100uf but to no avail. on HO i measure .4 volts (switching).

Seems to me there is still somthing wrong with th bootstrap charging any Ideas??
 

SgtWookie

Joined Jul 17, 2007
22,230
You ARE aware that the boost cap will discharge rather quickly, right? And that you can't leave the high-side MOSFET on for very long before you have to bring it back down again, and charge it with the low side MOSFET, right?
 

Thread Starter

ian123

Joined Aug 24, 2011
73
yes but i figure it should turn a motor on briefly jst so i can see it turn or am being stupid.

I caculated the cap to be 10 UF as per delta t=C x Delta v /15micro amps

that should give me more than 2 secs
 

SgtWookie

Joined Jul 17, 2007
22,230
What is the voltage across the boost cap?

Be careful, as if it turns on, you'd have about 300v on the low side of the boost cap.
 

Thread Starter

ian123

Joined Aug 24, 2011
73
Ok to test if my circuit works first I am running a 20vdc supply to the high floating side. the voltage at the boot cap is 4.65 volts
 

Thread Starter

ian123

Joined Aug 24, 2011
73
ok i found it i still have my diode connected to 5vdc on vdd should have it on vcc which is 12v

thanks for the help
 

SgtWookie

Joined Jul 17, 2007
22,230
If you don't have the boost cap diodes' anodes connected to +12v, the IR2110 will be in undervoltage lockout, and the high-side MOSFET will never turn on.
 

Thread Starter

ian123

Joined Aug 24, 2011
73
Hi sgt Wookie

mayb you cn help wih this as well the circuit works no at 22 volts on the floating side. I would like to draw tree ap rom thi circuit but dont seem to be able. th circuit locks up the pic has to be reset as i ramp up the voltage from 20 to 25 volts
 

Thread Starter

ian123

Joined Aug 24, 2011
73
sorry my keyboard is sticking i would like to draw 3 amps from this circuit at 300volts dc but it seems that at about 25v the circuit hangs up i then have to switch it off turn down the voltage to 20v then it works again
 

praondevou

Joined Jul 9, 2011
2,942
How far is the DC-bus capacitor from the mosfet bridge? It should be very near to it.

Are you saying the PIC locks up? This may be due to noise generated by your power circuit. That means short traces/wires, main cap near the bridge, the PIC not near the power circuit. There are other measures, but verify this first.
 
Top