Has any one used any of the International Rectifier reference boards? Look on thier web site www.irf.com for IRAUDAMP8
reference 4 channel development board, see what you think.
reference 4 channel development board, see what you think.
I have a couple of their ancient IRAUDAMP1 boards. I just recently finished a build with them. The sound is powerful and clean.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=145533
What kind of info are you looking for here besides "yeah, I have one..... I like it/don't like it"?
http://www.diyaudio.com/forums/showthread.php?s=&threadid=145533
What kind of info are you looking for here besides "yeah, I have one..... I like it/don't like it"?
If the IR power amp and PCB are of good quality in both sound and design then 120W X 4 for $294 is a good price. I have no
experience with class D amps other than what I read about them.
So yes any information on the sound of any of the IR reference boards would be appreciated. In looking at the reference design
I would think the B+B- supplies should be applied after the supplies are fully up i.e. with a delayed relay?
By the way love the capacitor bank.
thanks
experience with class D amps other than what I read about them.
So yes any information on the sound of any of the IR reference boards would be appreciated. In looking at the reference design
I would think the B+B- supplies should be applied after the supplies are fully up i.e. with a delayed relay?
By the way love the capacitor bank.
thanks
Have a look at the link below for a company that sells amps based on the IRS2092S. Check ebay for these as well; just search for "Class D Power Amplifier". They are capable of twice the power of the IRAUDAMP8 and two of them will still be cheaper.
I've been tempted to buy one of these, but the last thing I need is another amplifier board.
http://classdaudio.com (no affiliation)
I've been tempted to buy one of these, but the last thing I need is another amplifier board.
http://classdaudio.com (no affiliation)
I have experience with few models of IRAudamps. starting with Iraudamp1, after 4, 5, 7 and now i'm waiting to get Iraudamp8.
what i can say about them, is that a reference board is not designed to be used as an ready amplifier, is just for evaluating the parts used in that design, they are triyng to squeeze the performance out from the featured parts. that's why is some cases if you want to use them, you will discover that some stages are missing or another could be implemented better. another issue is that some reference designs (not only IRF) are made to promote some just released, expensive and hard to find parts, and if you don't have them available, you may discover that there are few other options arround to get similar performances with less cost and less effort.
i remade the design for Iraudamp7, to be able to deliver a wider range of power, from 50W to 500W per channel, using normal MOS-FET's transistors, which are more accessible and the user can choose the parts which will fit better for his design. i remade the layout, added some parts, shrink the board, made-it more symetric. you can see on my website this board.
regarding Iraudamp8, i'm not sure that is the best ideea to fit inside a single case 4 drivers. anyway, i'm about to try, let's see the result. from my experience, i can say that IRS2092S already has a significant temperature rise, esspecially when driving big gate charge MOS-FET's at hi sw. freq. putting 4 die IC's in a single case, will accentuate this. another issue is that the GND must be very well designed, otherwise is easy to get GND loop. a 4 layers pcb is the minimum requirement to avoid the loop and to get good EMI performaces. anyway, i think most of the designs which will use this ic will be in a power range under 200W per channel, and they will use Direct-FET MOS-FET's which are probably the best in the field, with lowest gate-charge, Rds-ON and switching timings. also the assembly technology required is not accesible for everybody, and this may be another issue against spreading this solution among DIY-ers.
what i can say about them, is that a reference board is not designed to be used as an ready amplifier, is just for evaluating the parts used in that design, they are triyng to squeeze the performance out from the featured parts. that's why is some cases if you want to use them, you will discover that some stages are missing or another could be implemented better. another issue is that some reference designs (not only IRF) are made to promote some just released, expensive and hard to find parts, and if you don't have them available, you may discover that there are few other options arround to get similar performances with less cost and less effort.
i remade the design for Iraudamp7, to be able to deliver a wider range of power, from 50W to 500W per channel, using normal MOS-FET's transistors, which are more accessible and the user can choose the parts which will fit better for his design. i remade the layout, added some parts, shrink the board, made-it more symetric. you can see on my website this board.
regarding Iraudamp8, i'm not sure that is the best ideea to fit inside a single case 4 drivers. anyway, i'm about to try, let's see the result. from my experience, i can say that IRS2092S already has a significant temperature rise, esspecially when driving big gate charge MOS-FET's at hi sw. freq. putting 4 die IC's in a single case, will accentuate this. another issue is that the GND must be very well designed, otherwise is easy to get GND loop. a 4 layers pcb is the minimum requirement to avoid the loop and to get good EMI performaces. anyway, i think most of the designs which will use this ic will be in a power range under 200W per channel, and they will use Direct-FET MOS-FET's which are probably the best in the field, with lowest gate-charge, Rds-ON and switching timings. also the assembly technology required is not accesible for everybody, and this may be another issue against spreading this solution among DIY-ers.
please read again carefully. i mentioned about IRS2093S. cannot route and fanout on single layer the pcb based on this IC with good EMI and electrical performances. and yes, indeed, my Iraudamp7 based amplifier is made on single layer pcb, same as the reference designs. anyway i learnt my lesson with this and i won't do-it again on single layer since the board price was almost the same as for two layers and doesn't worth all the effort to design and squeeze all the tracks on single layer.
cnx said:
Do you think this is wise (are there any significant drawbacks)?
... or would I be better off running the power supply down to 40V and possibly choosing a different amp module. I don't want to do it just because it is possible, but I'd like to make the most of what I have available.
Heatsinking is not a problem as this old amp has plenty of metal and space.
+-80V is about the maximum supply voltage which the IRS2092S can handle safely. don't forget the bus-pumping phenomenon. this can increase the supply rails voltage up to +-95V or even more. for this boards i advise to use max. +-72V to have a safety margin of at least 25V on each rail.
i can offer you TA0105 Amplifier which works wery well at this supply voltage. i just got some boards with this IC and they are able to handle high power. but for this you can contact me private, i don't want to be off-topic.
i can offer you TA0105 Amplifier which works wery well at this supply voltage. i just got some boards with this IC and they are able to handle high power. but for this you can contact me private, i don't want to be off-topic.
Eva said:Come on... You CAN'T do good and efficient class D on single layer PCBNo matter what you use to drive the MOSFET.
why cannot be done ??? what means good and efficient ?
this is quoted from the Iraudapm manual: "The design shows how to implement peripheral circuits on an optimum PCB layout
using a single sided board."
some measured performances:
at +-40V DC i got 87% efficiency on 4 Ohm load, 1KHz sine and IRF540N MOS-FET's 87.7% with IRF540Z. and this with the ~ 6W idle power consumption.
THD+N 0.034 at 150W measured A-weighted, 1KHz sine.
later i will post some photos of the measured data and PCB, close to the power stage.
Eva said:Come on... You CAN'T do good and efficient class D on single layer PCB
ha-ha, maybe cnx can't, I duno, but generally why no? Efficiency should be ok (with 540z we have 94% on the 2L PCB). Only EMI would be real problem, and of course 2 layer version will outperform single one yet.
Eva said:Come on... You CAN'T do good and efficient class D on single layer PCB
It's certainly not easy (and rather pointless anyway?), but so long as you provide good low-inductance current return paths (for *all* the switching currents) using coplanar structures it's not completely impossible.
Of course trying to do this for all the driver/gate/source/drain switching currents in a class-D output stage would probably drive you insane, and might turn out to be impossible in the end anyway -- so why not just use a multilayer PCB, they're cheap enough?
(unlike the customised 19-layer BGA package mounted on a 14-layer RF PCB that I'm working on right now...
Ian
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.