No problem Marc,
I'm glad that this is going forward.
Do you have an estimation on how big the PCB will be ?
Best Regards,
Savu
I'm glad that this is going forward.
Do you have an estimation on how big the PCB will be ?
Best Regards,
Savu
Hi Savu,
taken longer than expected to get parts into library, plus been at work this weekend.
Got parts in and double checked them and into the schematic, I expect most of it to be entered this evening. Gonna look at SMD alternatives to the zenors, I think there are some in SOT23 that have the same specs.
I did expect to be further so appologies.
taken longer than expected to get parts into library, plus been at work this weekend.
Got parts in and double checked them and into the schematic, I expect most of it to be entered this evening. Gonna look at SMD alternatives to the zenors, I think there are some in SOT23 that have the same specs.
I did expect to be further so appologies.
marc,
You don't have to apologize, i went to work this weekend too, i have lots of other important stuff. This is a project that we do for fun.
So in you're free time only.
Best regards,
Savu Silviu
You don't have to apologize, i went to work this weekend too, i have lots of other important stuff. This is a project that we do for fun.
So in you're free time only.
Best regards,
Savu Silviu
Hi Savu,
i sent an e-mail to the sinner_savu address from your profile last night !!!
Well into the schematic, wot I've done is attached.
I have some concern about how the LM311 is wired up, the output seems to go to the +3V supply?
If I get chance to night I'll get the rest of the schematic entered, then it will transfer straight to the PCB package. Had a look at some 3W surface mount resistors, they are rather large, so I think PTH may be best. Could do with a replacement for the TIP31C, something in a DPAK or D2PAK.
i sent an e-mail to the sinner_savu address from your profile last night !!!
Well into the schematic, wot I've done is attached.
I have some concern about how the LM311 is wired up, the output seems to go to the +3V supply?
If I get chance to night I'll get the rest of the schematic entered, then it will transfer straight to the PCB package. Had a look at some 3W surface mount resistors, they are rather large, so I think PTH may be best. Could do with a replacement for the TIP31C, something in a DPAK or D2PAK.
Hi Savu, Marc,
I am loving to see your interaction!
I would prefer the 5020 more than the 5015.
The Body diodes differ much more than you may think.
It is not just about trr, it is more about Qrr!
During hard switching situations (typically when the output current is higher than the ripple current of the filter) the biggest portion of the switching losses is caused by Qrr.
The small body diode of the 5020 should be sufficient for traditional class D concepts, because the body diode has to conduct the current only during the dead time.
Unfortunately the max. dv/dt after reverse recovery are not specified for both types.
Sometimes I am wondering about the inconsistent way of IR specifications!
BTW:
For IRFB4227 and 4127 max dv/dt is specified, but the Qrr of the body diode cannot compete with the Qrr of the 5020 at all, especially not at a di/dt of 500A/us.
Frankly speaking - I would love to get the 5020 with fully specified diode and TO220 package.
P.S.
As long as you can live with 100V devices - the IRF540Z is a nice one.
P.P.S.
No specific means for dead time adjustment in your circuit?
Do you plan to trim it by RF1, RF2, R5, R6 and/or supply voltage of the gate drivers?
I am loving to see your interaction!
I would prefer the 5020 more than the 5015.
The Body diodes differ much more than you may think.
It is not just about trr, it is more about Qrr!
During hard switching situations (typically when the output current is higher than the ripple current of the filter) the biggest portion of the switching losses is caused by Qrr.
The small body diode of the 5020 should be sufficient for traditional class D concepts, because the body diode has to conduct the current only during the dead time.
Unfortunately the max. dv/dt after reverse recovery are not specified for both types.
Sometimes I am wondering about the inconsistent way of IR specifications!
BTW:
For IRFB4227 and 4127 max dv/dt is specified, but the Qrr of the body diode cannot compete with the Qrr of the 5020 at all, especially not at a di/dt of 500A/us.
Frankly speaking - I would love to get the 5020 with fully specified diode and TO220 package.
P.S.
As long as you can live with 100V devices - the IRF540Z is a nice one.
P.P.S.
No specific means for dead time adjustment in your circuit?
Do you plan to trim it by RF1, RF2, R5, R6 and/or supply voltage of the gate drivers?
Hello Choco ...
I have both 5020 and 5015 sa i can test both.
The Schematic that I proposed is not mine but it is verified and functional, so i decided to do it even if it lacks protection dead time adjust and so on.
I want to use this mosfet because i want to see exactly how small can i make a class d amp.
This thread is also intended to show people that SMD's are not so hard to use and also in class d it has lots of benefits.
My only problem is that i can't make 4 layer board in my home so the design will be 2 layer only for the given moment.
Regards,
Savu Silviu
I have both 5020 and 5015 sa i can test both.
The Schematic that I proposed is not mine but it is verified and functional, so i decided to do it even if it lacks protection dead time adjust and so on.
I want to use this mosfet because i want to see exactly how small can i make a class d amp.
This thread is also intended to show people that SMD's are not so hard to use and also in class d it has lots of benefits.
My only problem is that i can't make 4 layer board in my home so the design will be 2 layer only for the given moment.
Regards,
Savu Silviu
I'm not an expert so ... i will try to make it as best i can ...
i will do oscilograms and also FFt reading and post ...
i'll post the result as we go .. but first i need to build it after Marc does the PCB design
i will do oscilograms and also FFt reading and post ...
i'll post the result as we go .. but first i need to build it after Marc does the PCB design
very interesting. I 
SMD and I mosfets. SMD has so many advantages and removes some more esoteric stuff; tho replaces it with freaky stuff 😉 I need to get myself a decent oven amd hot air rework station, for now i'm content hand soldering hundreds of 0805, 0603 and 0402 lol. 0402 was a bit insane.
I would like to construct myself an integrated bridged DAC with Class D amp into a sub. sealed is essential, (perfaps fan intake and exhaust, though could dissipation just be done in a way that placed the surfaces to be cooled on the outsiede and the rest around the inside with thermals??, as I have no real experience or the right equipment for building speakers, but I have a design in mind for that part, looking at the amp and dac part.
SMD and I mosfets. SMD has so many advantages and removes some more esoteric stuff; tho replaces it with freaky stuff 😉 I need to get myself a decent oven amd hot air rework station, for now i'm content hand soldering hundreds of 0805, 0603 and 0402 lol. 0402 was a bit insane.I would like to construct myself an integrated bridged DAC with Class D amp into a sub. sealed is essential, (perfaps fan intake and exhaust, though could dissipation just be done in a way that placed the surfaces to be cooled on the outsiede and the rest around the inside with thermals??, as I have no real experience or the right equipment for building speakers, but I have a design in mind for that part, looking at the amp and dac part.
Last edited:
yeah wouldnt wish 0402 by hand on anyone ;S 0603 is no problem though (flux, flux and more flux 😉
We do 0402 by hand (using a hot air gun) all the time at work, no problem.
0201 are an entirely different problem -- and then there's 01005...
http://www.murata.com/articles/ta06c2.pdf
-- in the photo at the bottom of the page, 0402 are the *huge* ones 🙂
Last edited:
Basic placement for the main circuitry, got to do the 12V supply, plus need to add output capacitor CG1.
Could do with an SMD replacement for theTIP31.
Gonna have a beer break then I do a bit of routing. So far there will be 3 links around the main circuitry! I want to keep the bottom layer as ground and ground only (at the moment).
The size below is 93 x 60 mm, though this will grow as we need to add some copper to remove the heat from the FETs.
Once we do this board, I think we should have a play at doing a double sided placement, 4 layer board, I reckon we could possibly half the size, but that can wait.
Using SMD power devices, brings some interesting and challenging assembly problems. In order to handle the heat and current the PCB has a mixture of Soldermask defined pads and non soldermask defined pads (I will dig an article out with pictures), this in turn means the soldermask defined pads require more heat to solder. In an oven you have to balance the temp so the solder on these pads reflows without drying the joints on the non solder mask defined pads. I belive using some sort of heat gun should suffice for home assembly, we shall find out.
Back soon, well after a can or 2.
Could do with an SMD replacement for theTIP31.
Gonna have a beer break then I do a bit of routing. So far there will be 3 links around the main circuitry! I want to keep the bottom layer as ground and ground only (at the moment).
The size below is 93 x 60 mm, though this will grow as we need to add some copper to remove the heat from the FETs.
Once we do this board, I think we should have a play at doing a double sided placement, 4 layer board, I reckon we could possibly half the size, but that can wait.
Using SMD power devices, brings some interesting and challenging assembly problems. In order to handle the heat and current the PCB has a mixture of Soldermask defined pads and non soldermask defined pads (I will dig an article out with pictures), this in turn means the soldermask defined pads require more heat to solder. In an oven you have to balance the temp so the solder on these pads reflows without drying the joints on the non solder mask defined pads. I belive using some sort of heat gun should suffice for home assembly, we shall find out.
Back soon, well after a can or 2.
actually I got some 0402 today and seem to remember this before,that are actually about the same size as 1206, strange. hot air gun is cheating 😉 i'm talking naked eye and hand solder or its a foul 😉 need to get a rework or just a decent hot air gun. I have one I use for heatshrink and another small butane one that does fine, but its borderline too hot and hard to control.
very keen on the 4 layer board, will see how this pans out. the dac i'm building at the moment os a mix of 2 layer and 4 layer smd and 4 layer sure allows some neat tricks
very keen on the 4 layer board, will see how this pans out. the dac i'm building at the moment os a mix of 2 layer and 4 layer smd and 4 layer sure allows some neat tricks
- Status
- Not open for further replies.