My old eyes still don't see the problem
I don't see any difference to the reference in post #26
I know I have mirrored the component in the schematic but it should not mean that it is mirrored in the pcb??
I just don't understand what you are trying to say, can you be more specific? Which pins should I swap?
look at the direction of the diode and where you have the pins labeled. You have pin 1 bottom right which is pin 2 in the device. Inbound to pin 2 will back bias the diode, no?
EDIT: Yes you have mirrored it in the schematic and yes I am worried about what happens to the board when you do that. Perhaps I am wrong. Less confusing to not do so.
EDIT: Yes you have mirrored it in the schematic and yes I am worried about what happens to the board when you do that. Perhaps I am wrong. Less confusing to not do so.
Last edited:
So, you think it is more clear like this:
An externally hosted image should be here but it was not working when we last tested it.
To me, it is a bit messy like this. All I did was changed the componentas not mirrored. In Eagle it makes no different if I mirror components in the schematic or not, it will be in on correct way on the board.
Still my question is whether there is incorrect wirings to wrong pins on vom1271 or not?
Then all ok. Apologies for the detour... (You could have wired to the upper one first and rotated the lower one. It would be a bit better. But if mirroring on the schematic isn't an issue then it doesn't matter. I just found it confusing at first.)
Back to the VOM1271 and the dual IRFB4115PBF. Am I interpreting the datasheets for each correctly if I say:
1. the IRFB4115PBF needs about 8V or more Vgs to be in full saturation (no increase in voltage at the gate will increase current from drain to source)
2. A current of 10mA to the VOM1271 produces 8.4V Vgs (15uA) for the IRFB4115PBF; a current of 20mA produces 8.7V
3. The PICAXE will supply 5V, 18mA
and hence the requirement is satisfied.
Do I have my terminology correct? Where in the data sheet is the maximum voltage for the VOM1271?
I note that this arrangement was discussed briefly in this thread
http://www.diyaudio.com/forums/solid-state/155939-mosfet-relays-9.html#post3453615
1. the IRFB4115PBF needs about 8V or more Vgs to be in full saturation (no increase in voltage at the gate will increase current from drain to source)
2. A current of 10mA to the VOM1271 produces 8.4V Vgs (15uA) for the IRFB4115PBF; a current of 20mA produces 8.7V
3. The PICAXE will supply 5V, 18mA
and hence the requirement is satisfied.
Do I have my terminology correct? Where in the data sheet is the maximum voltage for the VOM1271?
I note that this arrangement was discussed briefly in this thread
http://www.diyaudio.com/forums/solid-state/155939-mosfet-relays-9.html#post3453615
Last edited:
No, the other way around.
An externally hosted image should be here but it was not working when we last tested it.
BTW I doubt you need the 0.1uF cap (C45) as this bypass cap is only recommended if the reg is a fair distance from the voltage source.
You will also need to adjust the "power on" LED resistor (R43) to ensure the current is within the LED's limits. (Or remove these if you don't need the indicator.) But I guess you realised this already.
You will also need to adjust the "power on" LED resistor (R43) to ensure the current is within the LED's limits. (Or remove these if you don't need the indicator.) But I guess you realised this already.
Hopefully someone is able to answer this...
Meanwhile, I was confident that there would not be any major changes to the schematic, so I designed the board layout. I tried to make it as small as conveniently possible with mostly through-hole components.
Please note that there are no mounting holes. I decided to have the board fit into a 100mm U-channel alu-profile and mounted from the transistors. If you think this is not proper way to mount it I could replace the fuses with smaller ones and make room for mounting holes in the corners and probably one in the middle of the bigger capacitors.
Next, if nothing is to be changed, I will double check the layout have a small set of boards manufactured.
Here are the files:
board
schematic
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Post #85 is correct and if for some reason the proposed FET's wont open enough with 18mA through the VOM there are plenty of pin compatible FET's that will . The 18 mA output is available on each output pin of the picaxe ( with the 100ohm series resistor it will be between 18 and 21 mA depending on the forward voltage of the LED's in the VOM )
Max output voltage of the VOM can be read from the second graph in the datasheet .
PICAXE-08M2
Sourced or sunk through chip : 170mA
Sourced or sunk per pin : 25mA
Nice work on the PCB !
Yo could use the "smash" function in Eagle on your components to get your part numbers more visible .
Cheers ,
Rens
Last edited:
I just updated the other thread with the final schematic I modelled. Yes, you need the protection diodes and laying two in series with a 100nF cap to ground at their junction is meant to reduce coupling of high frequency noise. The LED is nothing but a power on indicator. It consumes current (small but some). I suspect in the context of your project the indicator is completely unnecessary. If you leave it in check the resistor to ensure you have the right current for the LED you choose. (I took it out of the modelling and I have yet to decide if I will have one in the next version of the little regulator.)
Nice work on the board. Did you route everything manually? I can't imagine how long it took you to layout and route the board.
How are you cooling your diodes?
(BTW you might find the discussion here interesting/helpful.)
Nice work on the board. Did you route everything manually? I can't imagine how long it took you to layout and route the board.
How are you cooling your diodes?
(BTW you might find the discussion here interesting/helpful.)
Last edited:
Nice work !
One more tip on smash ( see attachments ) if you right click on the large cross hair of the components name , select "move" and then keep right clicking and move your mouse until the "names" cross hairs are aligned , as I did with R17 and R32 .Gives a very Pro look
Could be wise to put some decoupling caps on IC3 rails , like 100nF , close to the chip ,with all the class D stuff in the same housing ?
Almost there !
Cheers ,
Rens
Attachments
Last edited:
I made the suggested changes by doctordata and SGK.
Here are the latest versions:
Eagle files ZIP
Schematic PDF
Board PDF
SGK, everything is done manually, because I wanted to have the layout as symmetric as possible and to have as wide traces as possible..
I'm going to cool the diodes (and fets) by installing the board inside a U-channel aluminium (100mm x 50mm x 4mm ~ 4" x 2" x 1/6"). The board is mounted from the component mounting holes and if the alu channel does not provide enough coolig, I will add a heatsink mounted to the U-channel on the oppisite side to the board. This would make a nice sides for the amplifier enclosure.
Here are the latest versions:
Eagle files ZIP
Schematic PDF
Board PDF
SGK, everything is done manually
, because I wanted to have the layout as symmetric as possible and to have as wide traces as possible..I'm going to cool the diodes (and fets) by installing the board inside a U-channel aluminium (100mm x 50mm x 4mm ~ 4" x 2" x 1/6"). The board is mounted from the component mounting holes and if the alu channel does not provide enough coolig, I will add a heatsink mounted to the U-channel on the oppisite side to the board. This would make a nice sides for the amplifier enclosure.
Last edited:
I made the suggested changes by doctordata and SGK.
Here are the latest versions:
Eagle files ZIP
Schematic PDF
Board PDF
SGK, everything is done manually
I'm going to cool the diodes (and fets) by installing the board inside a U-channel aluminium (100mm x 50mm x 4mm ~ 4" x 2" x 1/6"). The board is mounted from the component mounting holes and if the alu channel does not provide enough coolig, I will add a heatsink mounted to the U-channel on the oppisite side to the board. This would make a nice sides for the amplifier enclosure.
Good work mate !
Now let's see if it works
Cheers ,
Rens
SGK, in post #80 we were discussing about resistor values for 12V. It should have been 15V, my mistake. So, the resistor values for 15V should be e.g. 1210R and 243R, right?
I still decided to use LM317, or would you prefer to use LT1084 as in the other thread? The resistor values would be the same then for 15V?
BR, Sami
I still decided to use LM317, or would you prefer to use LT1084 as in the other thread? The resistor values would be the same then for 15V?
BR, Sami
Your traces look wide enough for me , have a look at
The CircuitCalculator.com Blog PCB Trace Width Calculator
Always wise to go for a 2 Oz copper board for power supplies with this kind of currents . Cooling won't be a problem ,Your U-channel will be sufficient .
I Always go for the manual tracing in Eagle , takes a bit of time , but so much satisfaction and much better results ! But you did a painstaking job !
I'll have an early night tonight , watching my home country playing soccer against Mexico tonight at 2 AM with the whole family , so will check all your posts tomorrow morning again !
Have a good one !
Rens
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.