Hi all,
I have mounted the new PM50 transformer on the amplifier and now I have a new probelm....
If I use only one channel on 4ohm load each channel gives me around 250Wrms and the supply rails falls only to around 47V
(they are regulated to 50V).
If I put it in bridge mode I can get only around 300Wrms on 4ohm then the amplifier mutes.
Even when the amplifier mutes the supply rails are still around 46V so I think that the problem now is not the transformer itself.
Please help, is there anything wrong in connecting two class D amps in brigded mode that are not synchronized? I don't understand why the IRS2092 goes in protection
thank you
I have mounted the new PM50 transformer on the amplifier and now I have a new probelm....
If I use only one channel on 4ohm load each channel gives me around 250Wrms and the supply rails falls only to around 47V
(they are regulated to 50V).
If I put it in bridge mode I can get only around 300Wrms on 4ohm then the amplifier mutes.
Even when the amplifier mutes the supply rails are still around 46V so I think that the problem now is not the transformer itself.
Please help, is there anything wrong in connecting two class D amps in brigded mode that are not synchronized? I don't understand why the IRS2092 goes in protection
thank you
Hi all,
I think now I have fixed my amp
The problem was the low side current limit settings.
When I was using the small RM14 trafo I noticed the amp going
in shutdown at full power but the rails were only 30V.
With the PM50 trafo the rails drops to 42V when the amps shuts down.
Now I am running with no current limit at the low side and with 35A at the high side. What is strange is that the low side current limit was set to 25A but it was tripping much before probably due
to spikes.
The IRS2092 has a blanking interval on the rising edge to avoid this, how could it be possible?
On top of that I have increased the dead time to 45ns (the maximum possible) and I have reduced the SMPS frequency to
70kHz.
I have measured around 700Wrms continuous and around 850Wrms with burst signals when running on a quite discharged car battery. The voltage on the battery drops to 10.2V at full power.
Is there anything wrong in my layout that can create voltage spikes that trips the current limit? The same behaviour is happening in stereo mode when driving 2ohm load.
I have discovered that the IRS2092 is basically an IRS20955 with the input op-amp, am I right? The information on the IRS20955 datasheet are much more complete than the one in IRS2092 datasheet expecially regarding the protections.
Now i have to try some resistors values to have an acceptable current limit on the low side mosfet, I don't want to risk to burn up everything if I short the output....
thank you
-marco
I think now I have fixed my amp
The problem was the low side current limit settings.
When I was using the small RM14 trafo I noticed the amp going
in shutdown at full power but the rails were only 30V.
With the PM50 trafo the rails drops to 42V when the amps shuts down.
Now I am running with no current limit at the low side and with 35A at the high side. What is strange is that the low side current limit was set to 25A but it was tripping much before probably due
to spikes.
The IRS2092 has a blanking interval on the rising edge to avoid this, how could it be possible?
On top of that I have increased the dead time to 45ns (the maximum possible) and I have reduced the SMPS frequency to
70kHz.
I have measured around 700Wrms continuous and around 850Wrms with burst signals when running on a quite discharged car battery. The voltage on the battery drops to 10.2V at full power.
Is there anything wrong in my layout that can create voltage spikes that trips the current limit? The same behaviour is happening in stereo mode when driving 2ohm load.
I have discovered that the IRS2092 is basically an IRS20955 with the input op-amp, am I right? The information on the IRS20955 datasheet are much more complete than the one in IRS2092 datasheet expecially regarding the protections.
Now i have to try some resistors values to have an acceptable current limit on the low side mosfet, I don't want to risk to burn up everything if I short the output....
thank you
-marco
Attachments
Hi all,
I have fixed the current limit and now the amp is working fine.
I have installed it in my car and it sounds very good and very powerful.
Now I have to do only some mechanical finishing to close the enclosure but all the electronics stuffs are done.
Thank you all for your support
-marco
I have fixed the current limit and now the amp is working fine.
I have installed it in my car and it sounds very good and very powerful.
Now I have to do only some mechanical finishing to close the enclosure but all the electronics stuffs are done.
Thank you all for your support
-marco
Hi Luka,
Of course as soon as I finish I will put some pictures.
Now I am working on a smaller, lower power version of that amp, with lower voltage rails and smaller SMPS trafo (maybe still my RM14 core rewound). The structure will remain the same, still using the IRS2092 and the same mosfets. I am planning for something around 100+100Wrms with no possibility to bridge it
I want to use the low power version in combination with this amp to drive the mid-high section of my car; it sounds too good to be used only for subwoofer...
BTW: by now I have measured only the power of my amp:
I got 700Wrms continuous on 4ohm, BTL and around 900Wrms
with short burst signals. I would like to measure THD, are there any easy method, maybe using the PC soundcard as a spectrum analyzer?
thanks
Of course as soon as I finish I will put some pictures.
Now I am working on a smaller, lower power version of that amp, with lower voltage rails and smaller SMPS trafo (maybe still my RM14 core rewound). The structure will remain the same, still using the IRS2092 and the same mosfets. I am planning for something around 100+100Wrms with no possibility to bridge it
I want to use the low power version in combination with this amp to drive the mid-high section of my car; it sounds too good to be used only for subwoofer...
BTW: by now I have measured only the power of my amp:
I got 700Wrms continuous on 4ohm, BTL and around 900Wrms
with short burst signals. I would like to measure THD, are there any easy method, maybe using the PC soundcard as a spectrum analyzer?
thanks
design Class-d based on IRS20955
I'm designing 2*150W and 2*250W Class-D with IR solution based on IRS20955, we design the Class-D schematic and PCB according to IRAUDAMP4 reference design and have completed the primary version PCB and are debugging the PCB. during the course of debugging, I happen to the following problem:
1) The CSD of the IRS20955 is always charging and discharging and there is no PWM output, that is to say, altrough i have no load, the IC is always OCP
I'm designing 2*150W and 2*250W Class-D with IR solution based on IRS20955, we design the Class-D schematic and PCB according to IRAUDAMP4 reference design and have completed the primary version PCB and are debugging the PCB. during the course of debugging, I happen to the following problem:
1) The CSD of the IRS20955 is always charging and discharging and there is no PWM output, that is to say, altrough i have no load, the IC is always OCP
I also happen the current limit problem
I'm designing 2*150W and 2*250W Class-D with IR solution based on IRS20955, we design the Class-D schematic and PCB according to IRAUDAMP4 reference design and have completed the primary version PCB and are debugging the PCB. during the course of debugging, I happen to the following problem:
1) if the mosfet is irf4019B and all parameters is same to the attachment, the CSD of the IRS20955 is always charging and discharging and there is no PWM output, that is to say, altrough i have no load, the IC is always OCP.
Please give you advice how to solve the problem.
2) if I want use mosfet FDP3682, how i set the relevent parameters about IRS20955 and can make the class-d work.
Hope to your reply as soon as possible.
My E-mail: lucy.ma@gpe-hkg.com
ponymxy@163.com
Thanks a lot!
I'm designing 2*150W and 2*250W Class-D with IR solution based on IRS20955, we design the Class-D schematic and PCB according to IRAUDAMP4 reference design and have completed the primary version PCB and are debugging the PCB. during the course of debugging, I happen to the following problem:
1) if the mosfet is irf4019B and all parameters is same to the attachment, the CSD of the IRS20955 is always charging and discharging and there is no PWM output, that is to say, altrough i have no load, the IC is always OCP.
Please give you advice how to solve the problem.
2) if I want use mosfet FDP3682, how i set the relevent parameters about IRS20955 and can make the class-d work.
Hope to your reply as soon as possible.
My E-mail: lucy.ma@gpe-hkg.com
ponymxy@163.com
Thanks a lot!
there will be DC when the output voltage is greater than 31.8V
I'm designing and 2*250W Class-D with IR solution based on IRS20955, we design the Class-D schematic and PCB according to IRAUDAMP4 reference design, the MOSFET used is IRFB4020 .
during the course of debugging, I happen to the following problem:
1) if the output voltage is greater than 31.6V, there will be DC voltage in output. if the output power is less than 31.6V, there is not DC voltage, what is the reason?
Hope every to give your advice.
Thanks a lot!
I'm designing and 2*250W Class-D with IR solution based on IRS20955, we design the Class-D schematic and PCB according to IRAUDAMP4 reference design, the MOSFET used is IRFB4020 .
during the course of debugging, I happen to the following problem:
1) if the output voltage is greater than 31.6V, there will be DC voltage in output. if the output power is less than 31.6V, there is not DC voltage, what is the reason?
Hope every to give your advice.
Thanks a lot!
IRS20855 Unstable
Interesting how this guy says that the IRS20955 is unstable. No facts to back up this statement
To date we have built about 15,000 class D amplifier channels using the IRS20955 using rails varying from +/-32v up to +/-85v. If you know how to use the chip it works well. If you know the limitations of the chip as a good designer, then it works well.
The IRS2092 is the same chip with the integrator built in BUT the 20955 is more versatile as you can use whatever inegrator chip you desire and then also change the feedabck network to post coil
Stephen Mantz
Zed Audio Corp
Interesting how this guy says that the IRS20955 is unstable. No facts to back up this statement
To date we have built about 15,000 class D amplifier channels using the IRS20955 using rails varying from +/-32v up to +/-85v. If you know how to use the chip it works well. If you know the limitations of the chip as a good designer, then it works well.
The IRS2092 is the same chip with the integrator built in BUT the 20955 is more versatile as you can use whatever inegrator chip you desire and then also change the feedabck network to post coil
Stephen Mantz
Zed Audio Corp
Re: IRS20855 Unstable
I don't know about irs20855. Initially I've tried 20955 with selfoscillating amp (similar UcD) and RM10 N87 coil, maybe 1.5 years before, everything was ok, but next I changed coil to micrometals material #2, and got the same problem, namely DC at the output if 95% mod.index. I've checked dV/dt isn't problem, and much less than datasheet limitation, but it seems trigger inside of 20955 sometimes just lose the edge and keep the same state, instead of switching, of course this is not protection event. If you use clocked class D topology, you'll never see such effect (DC latching on the output, as I remember negative rail), but only THD worse at 90-95% of modulation index, next I've tried 4 other drivers (3 IR and 1 ST) in the same layout without any problem.
MOER said:
I don't know about irs20855. Initially I've tried 20955 with selfoscillating amp (similar UcD) and RM10 N87 coil, maybe 1.5 years before, everything was ok, but next I changed coil to micrometals material #2, and got the same problem, namely DC at the output if 95% mod.index. I've checked dV/dt isn't problem, and much less than datasheet limitation, but it seems trigger inside of 20955 sometimes just lose the edge and keep the same state, instead of switching, of course this is not protection event. If you use clocked class D topology, you'll never see such effect (DC latching on the output, as I remember negative rail), but only THD worse at 90-95% of modulation index, next I've tried 4 other drivers (3 IR and 1 ST) in the same layout without any problem.
IRS20955 and 20922
The over current trip circuits are absolutely useless.
In all our lower powered ampslifier of less than 300w at 4 ohm we must set the current limit to 40 amps otherwise the chip keeps tripping.
We also use a PNP running off pin 2 (CSD) to send a signal to the on board microprocessor so the operator knows when the amplifier has a fault condition on the speaker lines.
Our new 600w 4 ohm amplifier, we shut down these upper and lower current sensors and dispensed with the PNP.
We use a current transformer, which was on board anyway for the microprocessor control, and the signal for this we use to shut down the amplifier. Works great.
Steve Mantz
Zed Audio Corp
zedaudio@aol.com
The over current trip circuits are absolutely useless.
In all our lower powered ampslifier of less than 300w at 4 ohm we must set the current limit to 40 amps otherwise the chip keeps tripping.
We also use a PNP running off pin 2 (CSD) to send a signal to the on board microprocessor so the operator knows when the amplifier has a fault condition on the speaker lines.
Our new 600w 4 ohm amplifier, we shut down these upper and lower current sensors and dispensed with the PNP.
We use a current transformer, which was on board anyway for the microprocessor control, and the signal for this we use to shut down the amplifier. Works great.
Steve Mantz
Zed Audio Corp
zedaudio@aol.com
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.