Replacement for IRFIZ24N
Hi!
Which MOSFETs are suitable for use with TAS5182 instead of IRFIZ24N? I can't obtain IRFIZ24N from my local suppliers. I have some FDD5612 from Fairchild. They have low gate charge and fast switching.
FDD5612:
Qg = 7.5 nC
td(on) = 8 ns
tr = 4 ns
td(off) = 24 ns
tf = 4 ns
Rds(on) = 55 mOhm
18A, 60V
A while ago I wrote to TI to ask them about dead time programming of TAS5182. They answered that there is no linear relationship between the resistor connected to DTC_HS and DTC_LS and dead time (tdtp), and they recommend to use the reference design from the datasheet. In the reference design from the datasheet with IRFIZ24N, DTC_HS and DTC_LS are connected to ground (Rdtp=0). If I use different MOSFETs how do I determine the value of resistors connected to DTC_HS and DTC_LS? Or should I just copy the reference design, and connect DTC_HS and DTC_LS to ground?
Regards,
Dejan
Hi!
Which MOSFETs are suitable for use with TAS5182 instead of IRFIZ24N? I can't obtain IRFIZ24N from my local suppliers. I have some FDD5612 from Fairchild. They have low gate charge and fast switching.
FDD5612:
Qg = 7.5 nC
td(on) = 8 ns
tr = 4 ns
td(off) = 24 ns
tf = 4 ns
Rds(on) = 55 mOhm
18A, 60V
A while ago I wrote to TI to ask them about dead time programming of TAS5182. They answered that there is no linear relationship between the resistor connected to DTC_HS and DTC_LS and dead time (tdtp), and they recommend to use the reference design from the datasheet. In the reference design from the datasheet with IRFIZ24N, DTC_HS and DTC_LS are connected to ground (Rdtp=0). If I use different MOSFETs how do I determine the value of resistors connected to DTC_HS and DTC_LS? Or should I just copy the reference design, and connect DTC_HS and DTC_LS to ground?
Regards,
Dejan
Hi Rookie
The design around the TAS5182 is limited by the maximum voltage of the TAS5182, therefore a change from 55V mosfets to 60V mosfets can NOT be recomended. The avalance breakdown of the mosfet is used to limit the peak voltage at the boot-strap node on the TAS5182.
Also it is recomended to copy as much as possible of the PCB layout. I can supply you with a copy of the gerber file if needed.
You can also use the non-isolated version of the IRFIZ24N, e.g. IRFZ24N - just remember to use isolation-washers betwen mosfet and heatsink.
Unfortunattely there is no direct replacement for the IR mosfets, as 55V mosfets are not standard.
You can find both types at www.farnell.com
rgds,
Kim
The design around the TAS5182 is limited by the maximum voltage of the TAS5182, therefore a change from 55V mosfets to 60V mosfets can NOT be recomended. The avalance breakdown of the mosfet is used to limit the peak voltage at the boot-strap node on the TAS5182.
Also it is recomended to copy as much as possible of the PCB layout. I can supply you with a copy of the gerber file if needed.
You can also use the non-isolated version of the IRFIZ24N, e.g. IRFZ24N - just remember to use isolation-washers betwen mosfet and heatsink.
Unfortunattely there is no direct replacement for the IR mosfets, as 55V mosfets are not standard.
You can find both types at www.farnell.com
rgds,
Kim
I found the IRFZ24N at a local supplier and it is very cheap! Only 0.5 EUR compared to 2.5 EUR at Farnell.
Kim, what do you think about filterless design? I've read the sloa023 (Reducing and Eliminating the Class-D Output Filter) and it seems that the filterless design provides less distrotion and higher fidelity. Also, BD modulation (with TAS5015) generates less EMI than AD modulation.
Regards,
Dejan
Kim, what do you think about filterless design? I've read the sloa023 (Reducing and Eliminating the Class-D Output Filter) and it seems that the filterless design provides less distrotion and higher fidelity. Also, BD modulation (with TAS5015) generates less EMI than AD modulation.
Regards,
Dejan
filterless design
Hi Rookie
The filterless design is targeted for applications where the speaker and amp are placed close together - like few cm.
Please note - it is not filterless - the cut-off frequencies are just at a higher frequency or it is only commonmode filter etc - you always need a filter
with a 8x150W system like the TAS5518-5182EVM you would never get it working.
for the same THD, BD-mode demands a better power supply than AD-mode.
Also the cross coupling between the half-bridges that is ocuring in BD-mode is a source for increased THD in the 1w range when using the TAS5182 with 40V supply, look at the graphs in the TAS5076-5182EVM apps note on www.ti.com.
rgds,
kim
Hi Rookie
The filterless design is targeted for applications where the speaker and amp are placed close together - like few cm.
Please note - it is not filterless - the cut-off frequencies are just at a higher frequency or it is only commonmode filter etc - you always need a filter
with a 8x150W system like the TAS5518-5182EVM you would never get it working.
for the same THD, BD-mode demands a better power supply than AD-mode.
Also the cross coupling between the half-bridges that is ocuring in BD-mode is a source for increased THD in the 1w range when using the TAS5182 with 40V supply, look at the graphs in the TAS5076-5182EVM apps note on www.ti.com.
rgds,
kim
Has anyone come across a clss D amp. genuinely delivering less than 0.01% THD+N across the audio band?? 
http://www.stereophile.com/solidpoweramps/805cia/index4.html
http://www.stereophile.com/solidpoweramps/805cia/index4.html
The TacT amp Millenium is rated at THD < 0.1% 20-20kHz, all powers up to 150W 8R, 300W 4R, although it's closer to 0.01-0.03% under most conditions.
My third generation PowerDAC (PowerDAC 3 prototype) is an open loop all-digital design - ie like the TacT Millenium - which weighs in at THD 0.02% - 0.03%, 20-20kHz, all powers up to 175W 8 ohms, 350W 4 ohms.
John Westlake has described on this forum his all-digital amps that give 120W 8R and THD 0.000x% (my apologies to John if I left out some zeros there) . This is a closed loop design
So yes, it's possible. Whether or not the latter two develop-ments result in commercially viable products remains to be seen.
Cheers
My third generation PowerDAC (PowerDAC 3 prototype) is an open loop all-digital design - ie like the TacT Millenium - which weighs in at THD 0.02% - 0.03%, 20-20kHz, all powers up to 175W 8 ohms, 350W 4 ohms.
John Westlake has described on this forum his all-digital amps that give 120W 8R and THD 0.000x% (my apologies to John if I left out some zeros there) . This is a closed loop design
So yes, it's possible. Whether or not the latter two develop-ments result in commercially viable products remains to be seen.
Cheers
Lars Clausen said:I measured this with one of my new designs. However i am not using the best ferrites yet, i will get them only in a few days.
Then i expect the lift at 10kHz will disappear.
Folks, thanks for your input...
I would appreciate some discussion of measurement conditions...in particular, the measurement bandwidth....
It is noteworthy that Atkinson (Stereophile), has yet to find a commercial class D amp. (digital or otherwise) with THD+N consistently below .01%
IVX: Absolutely agree with you. But now i am waiting for one of the ferrite types with very low THD. However the new amplifier allows including the choke in the feedback loop. And what happened was the THD became stable and frequency independent, but at a higher level than here. Something like 0.007%.
Anyway i am also on the trace of another contributing factor to improve the THD. So in the next week or so, i will have something that doesn't look so bad compared to other available modules.
However i dont want to comment on the 'Commercial available amplifier' discussion going on, since for me as a grown DIY'er these amplifiers have no special interest for me. Especially since the best ones you can get today is on the DIY market.
Anyway i am also on the trace of another contributing factor to improve the THD. So in the next week or so, i will have something that doesn't look so bad compared to other available modules.
However i dont want to comment on the 'Commercial available amplifier' discussion going on, since for me as a grown DIY'er these amplifiers have no special interest for me. Especially since the best ones you can get today is on the DIY market.
It can be found here: www.ymec.com, you can download 30 days demo for free.
And this is the sound card i am using in these tests:
http://www.digit-life.com/articles2/esi-julia/index.html
And this is the sound card i am using in these tests:
http://www.digit-life.com/articles2/esi-julia/index.html
IVX: Sorry to hear that, but there are many sofwares out there that does the same.
Example :
Spectra Labs: http://www.telebyte.com/pioneer/
Rightmark: http://audio.rightmark.org/index_new.shtml
and others
Example :
Spectra Labs: http://www.telebyte.com/pioneer/
Rightmark: http://audio.rightmark.org/index_new.shtml
and others
0.01% THD
Hi All
Low THD is not nessesarry a measure for good sound - but none the less attached is a graph showing just that with a non feedback design.
modulator TAS5518
power stage TAS5182, mosfet FDD5612 fairchild, used in PBTL (2channels is parrallel mode)
THD is qiute low considering the non feedback design - and this aross a large frequency range.
blue trace 1W, green trace 10W, red trace 80W
Measurement system, Audio Precesion system 2 - cascade with AES17 filter. Load 8R.
rgds,
Kim
Hi All
Low THD is not nessesarry a measure for good sound - but none the less attached is a graph showing just that with a non feedback design.
modulator TAS5518
power stage TAS5182, mosfet FDD5612 fairchild, used in PBTL (2channels is parrallel mode)
THD is qiute low considering the non feedback design - and this aross a large frequency range.
blue trace 1W, green trace 10W, red trace 80W
Measurement system, Audio Precesion system 2 - cascade with AES17 filter. Load 8R.
rgds,
Kim
Open loop digital systems are a big no-no, a stupid dead-end 
Power Supply Rejection Ratio is 0dB in these systems!!
On the other hand, closed loop digital systems are equivalent to any analog-controlled class D amplifier, but requiring fast, precise and expensive ADCs instead of standard DACs, so they are another dead end.
Power Supply Rejection Ratio is 0dB in these systems!!
On the other hand, closed loop digital systems are equivalent to any analog-controlled class D amplifier, but requiring fast, precise and expensive ADCs instead of standard DACs, so they are another dead end.
Kims: very interesting!
I have a couple of questions:
1..Where is the bottleneck in this setup? Can it be further improved, to even lower levels of THD?
2..How does the output inductor affect the THD figures at different loads?
3..How do you keep the supply rails stable enough to reach the levels of THD you are showing here?
Best regards
Lars Clausen
I have a couple of questions:
1..Where is the bottleneck in this setup? Can it be further improved, to even lower levels of THD?
2..How does the output inductor affect the THD figures at different loads?
3..How do you keep the supply rails stable enough to reach the levels of THD you are showing here?
Best regards
Lars Clausen
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.