I'd like to know if anyone have built it, and if the PCB I've started to design (first PCB design attempt) would work any good. I am worried that the tracks to gate resistors are too long. I plan to use no tracks for PSU wires, but directly solder them from top of PCB, to mosfet legs with bypassing, and a bit higher voltage rails for driver stage. Why single PCB? For home etching. Do not pay attention to using IRFP240 model instead of P-Channel part (couldn't find TO-247BH package for P-Channel) lack of connectors and such a stuff, I'd just like to know if the PCB tracks are short enough. There is a 10mA current source missing currently, and mute resistor, which I plan to solder right at IC legs. I am not sure if it is ok to post a schematic here, as it is from Bob's book. Basically it is a LME49830 circuit using LT1166 bias IC instead of Vbe multiplier.
To sum it all up, here are my questions: Anyone built it?
Are length of tracks ok?
Cheers, Vic.
To sum it all up, here are my questions: Anyone built it?
Are length of tracks ok?
Cheers, Vic.
Attachments
Last edited:
LT1166+LME49830 design by Bob Cordell
Hi Vic,
Take a look at my layout in this thread,
LME49830+LatFET, yet another for design review.
there is an optimal position for the LME49830 which will have short gate drive leads. The LME49830 heatsink is a limiting factor in reducing the gate drive track length. I even tried to match the gate drive lengths, but these are not a timing critical signals, so not really necessary.
If using a single sided PCB, I believe that there will be compromizes to the layout. Large area fills are required for supplies, grounds & the speaker output signal, due to the very high currents involved. Can't see how this can be done properly using only a Single-Sided PCB.
OPC's(Owen) "Wire" design, used 3 layers, but I am not convinced that it is really necessary, if you do a good 2-layer layout. AndrewT second that point, but I have not seen any of his designs posted for reviews.
I am sure Bob's design is sound. I do not have his book. Maybe Santa will give me one this xmas. I assume that there was no layout/gerber data for his design. Layout is critical in achieving maximum performance.
I thought of using the IRF MOSFETs originally, but the Lateral FET's look like they are a better choice. One good thing about the IRF FET's is that they are cheap and readilly available, so in that sense, they are a good choice. Can't recall if the Toshiba FET's have the same pinout, if they do then, this gives one more choices.
Good Luck with your project!!
Rick
Hi Vic,
Take a look at my layout in this thread,
LME49830+LatFET, yet another for design review.
there is an optimal position for the LME49830 which will have short gate drive leads. The LME49830 heatsink is a limiting factor in reducing the gate drive track length. I even tried to match the gate drive lengths, but these are not a timing critical signals, so not really necessary.
If using a single sided PCB, I believe that there will be compromizes to the layout. Large area fills are required for supplies, grounds & the speaker output signal, due to the very high currents involved. Can't see how this can be done properly using only a Single-Sided PCB.
OPC's(Owen) "Wire" design, used 3 layers, but I am not convinced that it is really necessary, if you do a good 2-layer layout. AndrewT second that point, but I have not seen any of his designs posted for reviews.
I am sure Bob's design is sound. I do not have his book. Maybe Santa will give me one this xmas. I assume that there was no layout/gerber data for his design. Layout is critical in achieving maximum performance.
I thought of using the IRF MOSFETs originally, but the Lateral FET's look like they are a better choice. One good thing about the IRF FET's is that they are cheap and readilly available, so in that sense, they are a good choice. Can't recall if the Toshiba FET's have the same pinout, if they do then, this gives one more choices.
Good Luck with your project!!
Rick
Hey, Rick.
I already took a look at your thread. I've made a LME49830 amp with point to point method, no hum and no noise. I also tried lateral FETs, but IRFPs sounded better to my ears, no idea why. In Bob's book there was no PCB design for this LME49830 and LT1166 combination. My idea is not to use a PCB tracks for power supply at all, by using wire connection soldered directly to FETs. I guess my only choice is to build it, and then I'll see how it works. Good luck with LatFET project as well.
Cheers, Vic.
I already took a look at your thread. I've made a LME49830 amp with point to point method, no hum and no noise. I also tried lateral FETs, but IRFPs sounded better to my ears, no idea why. In Bob's book there was no PCB design for this LME49830 and LT1166 combination. My idea is not to use a PCB tracks for power supply at all, by using wire connection soldered directly to FETs. I guess my only choice is to build it, and then I'll see how it works. Good luck with LatFET project as well.
Cheers, Vic.
Hi Vic,
Nice to hear about your successful experiences with these parts A bit surprised that you like the sound of the IRFP over the LatFet. I which I could hear the 2 versions at this time, before I commit to building anything. Good news, as the IRFP parts are certainly cheaper and more available.Which LatFets have you tried?
How about your heatsink choices ,to cool the LME?
Do you have a schematic, that I can look at?
Thanks
Rick
Nice to hear about your successful experiences with these parts A bit surprised that you like the sound of the IRFP over the LatFet. I which I could hear the 2 versions at this time, before I commit to building anything. Good news, as the IRFP parts are certainly cheaper and more available.Which LatFets have you tried?
How about your heatsink choices ,to cool the LME?
Do you have a schematic, that I can look at?
Thanks
Rick
The LT1166 looks interesting, did not know it existed, until now.
LT1166 only seems to be available from the LT web site, which is good to know.
Your comp placement looks good. Maybe separate the gate drive signals from the source outputs by 5mm or more (stretch the resistor lead lengths a bit) & see if you can get some ground plane between them!!. Not really sure if gate protection diodes are required?
Rick
LT1166 only seems to be available from the LT web site, which is good to know.
Your comp placement looks good. Maybe separate the gate drive signals from the source outputs by 5mm or more (stretch the resistor lead lengths a bit) & see if you can get some ground plane between them!!. Not really sure if gate protection diodes are required?
Rick
Thanks for suggestions, Rick. What I like about LT1166 - bias is temperature independent. I never use any protection diodes with IRFPs in audio. None have failed yet, also I believe I am pretty discharged all the time, cause I never have used any precautions to handle mosfets. For more expensive parts I do 
Cheers, Vic.
Cheers, Vic.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.