IRS2092 on breadboard - Page 3

ssanmor · 3rd April 2009, 02:59 PM

If you disconnect the mosfets you won't have a square wave at the output that is fed back to the input and produces the oscillation, so it won't work at all, to put it simple.
Perhaps you can inject a clock signal (see IRAUDAMP5 / 7 designs for details on how to do that), in order to force it to oscillate at its frequency.

rob mullen · 6th April 2009, 02:02 PM

ok,but what type of signal should i expect on pins HO and LO if the MOSFETS are connected and the feedback resistor (47K). a square wave of what amplitude? Also the rails are +20V and -20V, should i increase them?

ssanmor · 6th April 2009, 03:03 PM

Ok.
Don't really want to discourage you, in fact we are here to help you, but I sincerely think that you should read the datasheet of the chip first and get the necessary equipment.
You should get around 12Vpp square waves (50% duty cycle with no input) at LO-COM, and the same at HO-VS (although this one is more difficult to measure, as discussed previously here). For that, you need that Vcc-COM is correct (12V approx, and for your measurements it doesn't seem correct), and that Vb-Vs=12V as well (correct bootstrap voltage).

First, determine if you have 12V between Vcc and COM (pins 12 and 10).

rob mullen · 7th April 2009, 02:00 PM

ok i have checked Vcc with respect to COM and i am getting 12V. I also checked Vb-Vs and it is 0V

rob mullen · 7th April 2009, 02:18 PM

ok there is something up with the bootstrap voltage. i think i need to change the value of Rcharge. i need to lower it in order for Icharge>Iqbs

ssanmor · 7th April 2009, 03:21 PM

Ok, it is possible that with such low rails voltage the bootstrap capacitor is not precharging properly, try with a lower value and post on the results. However, it cannot be too low or it will contribute to discharge the capacitor during normal operation, limiting your max duty cycle and producing problems during clipping, for example.

Eva · 8th April 2009, 12:47 AM

Haven't you looked carefully at the datasheet? Worst case high side quiescent current is 1mA. The resistor should provide that current even with your low supply rails, and at least until Vbs has charged to approx 10V.

A current return path has to be provided too, the speaker usually does the job but a resistor to the lower rail will help to allow the amp to start with no load. Such low supply rails are a difficult situation. A series diode may help to prevent discharge as mentioned.

rob mullen · 8th April 2009, 07:43 PM

I have looked at the datasheet, i have the datasheet and some of the application notes regarding the IRS2092. i have been working off the application circuit that i uploaded a picture of previously. I took a look over the current protection descriptions in the application note AN-1138 and there is something that is confusing me. On page 10 of the application note AN-1138, Figure 13 is a diagram of the Bi-directional over current protection. why on that diagram is the resistor (Rcharge) omitted ? Rcharge that should be from VB (cathode side of Dbs) to the positive supply rail.

I connected a load to the amplifier today. I used a rheostat to simulate a speaker (8ohm load). the MOSFET on HO started to heat up very quickly, so i tuned it off.

Eva · 8th April 2009, 11:03 PM

Since this resistor is not related to current limiting, they may have just forgotten to include it.

If this is your first attempt with International Rectifier's integrated hi and lo side MOSFET drivers, you have made a bad choice. You should have started with a simpler IC like IR2010 or IR2110, so that level shifting and the rest of the stuff is outside the IC and you can gain control (and understanding) over it. The bad thing about self oscillating amplifiers is that they appear dead if anything is wrong.

The project in which I'm currently working is a big self-oscillating amplifier (>3KW). It uses IR2110. There is a PIC that can gain control over the modulator and drive the output stage. I used it at first as a square wave generator for testing the power stage. Now it works fine but this test mode is still in the code

If you don't understand the basics first, you won't be able to make IRS2092 work. For example, you say that the high side MOSFET got hot but it would seem that you are expecting us to figure out why for you. We can't (unless we have the prototype, a scope and a multimeter in front of us, like you may probably have). You should measure and interpret the data to find out (every voltage and current is telling you something).

This is not like class AB (where you should gain experience first too if you don't have any). Bear in mind that in switching amplifiers there are things happening up to 100Mhz and all the layout and component parasitics (resonances) will be excited. Learning to handle parasitics is a subject on itself.

Current limiting by Rds-on sensing (as done in IRS2092) is something that you should ignore too if this is your first attempt. This protection is very prone to false tripping if layout and switching is not optimum. It's based on parasitic voltage drops, and you may have plenty in a breadboard...

Do you have any experience in switching? Making an unregulated SMPS with IR integrated drivers may help to understand things too, since you wouldn't have to worry about the modulator, you would just have a fixed "signal generator" feeding the driver IC.

rob mullen · 9th April 2009, 08:58 AM

Yeah your right i should have started with a simpler IC, the IRS2092 was the first one i came across and i thought it was neat for all the features that included. however yes these features are proving to be troublesome. there is alot of conditions that need to be met in order for it to work properly, and i know building it on a breadboard is not ideal. I do want to design a PCB for this,but it would be nice if i could have it working on the breadboard first. With the PCB design there is alot to consider regarding grounding planes and such. i don't want to give up yet with this.

8th April 2009, 12:47 AM	#27
Eva diyAudio Member Join Date: Oct 2003 Location: Near the sea	Haven't you looked carefully at the datasheet? Worst case high side quiescent current is 1mA. The resistor should provide that current even with your low supply rails, and at least until Vbs has charged to approx 10V. A current return path has to be provided too, the speaker usually does the job but a resistor to the lower rail will help to allow the amp to start with no load. Such low supply rails are a difficult situation. A series diode may help to prevent discharge as mentioned. __________________ I use to feel like the small child in The Emperor's New Clothes tale

8th April 2009, 11:03 PM	#29
Eva diyAudio Member Join Date: Oct 2003 Location: Near the sea	Since this resistor is not related to current limiting, they may have just forgotten to include it. If this is your first attempt with International Rectifier's integrated hi and lo side MOSFET drivers, you have made a bad choice. You should have started with a simpler IC like IR2010 or IR2110, so that level shifting and the rest of the stuff is outside the IC and you can gain control (and understanding) over it. The bad thing about self oscillating amplifiers is that they appear dead if anything is wrong. The project in which I'm currently working is a big self-oscillating amplifier (>3KW). It uses IR2110. There is a PIC that can gain control over the modulator and drive the output stage. I used it at first as a square wave generator for testing the power stage. Now it works fine but this test mode is still in the code If you don't understand the basics first, you won't be able to make IRS2092 work. For example, you say that the high side MOSFET got hot but it would seem that you are expecting us to figure out why for you. We can't (unless we have the prototype, a scope and a multimeter in front of us, like you may probably have). You should measure and interpret the data to find out (every voltage and current is telling you something). This is not like class AB (where you should gain experience first too if you don't have any). Bear in mind that in switching amplifiers there are things happening up to 100Mhz and all the layout and component parasitics (resonances) will be excited. Learning to handle parasitics is a subject on itself. Current limiting by Rds-on sensing (as done in IRS2092) is something that you should ignore too if this is your first attempt. This protection is very prone to false tripping if layout and switching is not optimum. It's based on parasitic voltage drops, and you may have plenty in a breadboard... Do you have any experience in switching? Making an unregulated SMPS with IR integrated drivers may help to understand things too, since you wouldn't have to worry about the modulator, you would just have a fixed "signal generator" feeding the driver IC. __________________ I use to feel like the small child in The Emperor's New Clothes tale

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
IRS2092 customizable	DarkOne	Class D	28	12th December 2012 05:13 AM
Class-D Amp with IRS2092	mag	Class D	185	24th February 2012 03:35 PM
Another amp with IRS2092	mag	Class D	23	18th June 2010 07:25 PM
IRS2092 Amplifier	zox2003	Class D	22	31st August 2009 09:49 PM
Any suggestion for IRS2092	ja2dhc	Class D	4	2nd August 2008 10:53 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

3rd April 2009, 02:59 PM	#21
ssanmor diyAudio Member Join Date: Jan 2002 Location: Madrid	If you disconnect the mosfets you won't have a square wave at the output that is fed back to the input and produces the oscillation, so it won't work at all, to put it simple. Perhaps you can inject a clock signal (see IRAUDAMP5 / 7 designs for details on how to do that), in order to force it to oscillate at its frequency.

6th April 2009, 02:02 PM	#22
rob mullen diyAudio Member Join Date: Oct 2008	ok,but what type of signal should i expect on pins HO and LO if the MOSFETS are connected and the feedback resistor (47K). a square wave of what amplitude? Also the rails are +20V and -20V, should i increase them?

6th April 2009, 03:03 PM	#23
ssanmor diyAudio Member Join Date: Jan 2002 Location: Madrid	Ok. Don't really want to discourage you, in fact we are here to help you, but I sincerely think that you should read the datasheet of the chip first and get the necessary equipment. You should get around 12Vpp square waves (50% duty cycle with no input) at LO-COM, and the same at HO-VS (although this one is more difficult to measure, as discussed previously here). For that, you need that Vcc-COM is correct (12V approx, and for your measurements it doesn't seem correct), and that Vb-Vs=12V as well (correct bootstrap voltage). First, determine if you have 12V between Vcc and COM (pins 12 and 10).

7th April 2009, 02:00 PM	#24
rob mullen diyAudio Member Join Date: Oct 2008	ok i have checked Vcc with respect to COM and i am getting 12V. I also checked Vb-Vs and it is 0V

7th April 2009, 02:18 PM	#25
rob mullen diyAudio Member Join Date: Oct 2008	ok there is something up with the bootstrap voltage. i think i need to change the value of Rcharge. i need to lower it in order for Icharge>Iqbs

7th April 2009, 03:21 PM	#26
ssanmor diyAudio Member Join Date: Jan 2002 Location: Madrid	Ok, it is possible that with such low rails voltage the bootstrap capacitor is not precharging properly, try with a lower value and post on the results. However, it cannot be too low or it will contribute to discharge the capacitor during normal operation, limiting your max duty cycle and producing problems during clipping, for example.

8th April 2009, 07:43 PM	#28
rob mullen diyAudio Member Join Date: Oct 2008	I have looked at the datasheet, i have the datasheet and some of the application notes regarding the IRS2092. i have been working off the application circuit that i uploaded a picture of previously. I took a look over the current protection descriptions in the application note AN-1138 and there is something that is confusing me. On page 10 of the application note AN-1138, Figure 13 is a diagram of the Bi-directional over current protection. why on that diagram is the resistor (Rcharge) omitted ? Rcharge that should be from VB (cathode side of Dbs) to the positive supply rail. I connected a load to the amplifier today. I used a rheostat to simulate a speaker (8ohm load). the MOSFET on HO started to heat up very quickly, so i tuned it off.

9th April 2009, 08:58 AM	#30
rob mullen diyAudio Member Join Date: Oct 2008	Yeah your right i should have started with a simpler IC, the IRS2092 was the first one i came across and i thought it was neat for all the features that included. however yes these features are proving to be troublesome. there is alot of conditions that need to be met in order for it to work properly, and i know building it on a breadboard is not ideal. I do want to design a PCB for this,but it would be nice if i could have it working on the breadboard first. With the PCB design there is alot to consider regarding grounding planes and such. i don't want to give up yet with this.