48v on V_Bed by replacing SMD components?



  • I am exploring the possibility of driving 48V on the V_bed rail on a Duet 2 Ethernet. From what I see in the schematics the components that would need replacing are TR2, D5/6, and C91

    0_1555865364191_0cc5ccd5-71ea-4f73-a405-fd677b6c5598-image.png

    For TR2 a 60V mosfet could be used instead, such as a IPD031N06L3 G

    For D5 I'm struggling to determine what the allowable forward voltage is. The worst case I found was 30V for SOD-523F package, ie RB520S30. But it could also be 75V if it was a 1N4448WT

    Looking at a close up of the SMD it appears to have an E2 marking, which makes me think it's a 75V version
    0_1555871021518_070aacf9-3899-4c7b-9665-0af5cd8cd152-image.png
    I'm also thinking the LED is going to be affected, meaning replacing R65 with an 10k resistor.

    For C91 I'm thinking this is easy enough to just replace with another 10uf capacitor (what appears to be a 1206 package) with a higher voltage such as this: CL31A106KBHNNNE

    Does this all make sense? Am I missing something obvious?

    Granted this is all easy to do on paper, and not something very easily done on an existing board. The first step is just figuring out the circuit, then worrying about how to actually do it.



  • Why not just use a SSR? They're ~$20.



  • I have been using an SSR but it limits the operating frequency and feels unneeded. The SSR being used right now is this one from automation direct

    The point of not using an SSR would be to achieve a higher frequency than 50-60Hz, which is what the data sheet lists as the operating frequency for them.

    This would also be a way to save costs by not requiring a separate circuit to drive a 48v device, either by an SSR or the mosfet selection I'm proposing.



  • @mkelly said in 48v on V_Bed by replacing SMD components?:

    I have been using an SSR but it limits the operating frequency and feels unneeded. The SSR being used right now is this one from automation direct

    The point of not using an SSR would be to achieve a higher frequency than 50-60Hz, which is what the data sheet lists as the operating frequency for them.

    This would also be a way to save costs by not requiring a separate circuit to drive a 48v device, either by an SSR or the mosfet selection I'm proposing.

    I do not understand your concerns.

    For a bed heater the frequency range you listed is fine. Bed heaters rated for line voltage (110/220 VAC) are usually more powerful which is a good thing.

    Since the SSR connects directly to the bed heater output on the Duet it is about as simple as it can be.

    What do think you would improve by doing what you propose?

    Thanks.

    Frederick



  • @fcwilt

    @fcwilt said in 48v on V_Bed by replacing SMD components?:

    @mkelly said in 48v on V_Bed by replacing SMD components?:

    I have been using an SSR but it limits the operating frequency and feels unneeded. The SSR being used right now is this one from automation direct

    The point of not using an SSR would be to achieve a higher frequency than 50-60Hz, which is what the data sheet lists as the operating frequency for them.

    This would also be a way to save costs by not requiring a separate circuit to drive a 48v device, either by an SSR or the mosfet selection I'm proposing.

    I do not understand your concerns.

    For a bed heater the frequency range you listed is fine. Bed heaters rated for line voltage (110/220 VAC) are usually more powerful which is a good thing.

    Since the SSR connects directly to the bed heater output on the Duet it is about as simple as it can be.

    What do think you would improve by doing what you propose?

    Thanks.

    Frederick

    Hi Fred. I never said I wanted to be driving a bed heater from the V_bed line, but rather I wanted to drive 48v at a higher frequency (ideally 1-2khz range). This is an exploration at the possibility.



  • @mkelly The 50-60Hz is for AC SSRs. You linked to a DC SSR, so presumably you are using a DC SSR. So, there's no 50-60Hz limit in your application.

    I suspect you will spend considerably more than $20 going down this road, once time cost is factored in. Having said that, your analysis of what parts to replace is reasonable.



  • @mkelly said in 48v on V_Bed by replacing SMD components?:

    @fcwilt

    @fcwilt said in 48v on V_Bed by replacing SMD components?:

    @mkelly said in 48v on V_Bed by replacing SMD components?:

    I have been using an SSR but it limits the operating frequency and feels unneeded. The SSR being used right now is this one from automation direct

    The point of not using an SSR would be to achieve a higher frequency than 50-60Hz, which is what the data sheet lists as the operating frequency for them.

    This would also be a way to save costs by not requiring a separate circuit to drive a 48v device, either by an SSR or the mosfet selection I'm proposing.

    I do not understand your concerns.

    For a bed heater the frequency range you listed is fine. Bed heaters rated for line voltage (110/220 VAC) are usually more powerful which is a good thing.

    Since the SSR connects directly to the bed heater output on the Duet it is about as simple as it can be.

    What do think you would improve by doing what you propose?

    Thanks.

    Frederick

    Hi Fred. I never said I wanted to be driving a bed heater from the V_bed line, but rather I wanted to drive 48v at a higher frequency (ideally 1-2khz range). This is an exploration at the possibility.

    I understand but I still wonder why. There is nothing to be gained by driving a bed heater at those frequencies.

    What do you think you will gain versus a line voltage bed heater driven by an SSR connected to the Duet bed heater output?

    Frederick



  • @fcwilt Frederick. If you read what the OP said very carefully, you'll see that he never mentioned driving a bed heater, but rather using the V_bed line to drive some sort of 48V device.



  • @mkelly, a few thoughts about your original post.

    1. Why do you need to feed the 48V to the Duet, can't you just feed it to the + side of your load and letting TR2 close the path to ground?

    2. D5, D6 protecting each other against reversed voltage so I think they should be fine.

    3. If you leave V_BED as is (e.g. connected to +12 or +24V), the led ON current should be the same so no need to increase the resistor. In the OFF state of TR2, R65 will see 48V-V_BED so if you run on 12V, make sure it can sustain the increased power dissipation.

    4. You will need to replace TR2 since it can handle only 40V when open. Make sure to pick a compatible one, mostly rON low enough at the gate voltage that the duet provide.

    5. I used Auberins SRDD100 SSR with the default PWM frequency (very low, you can see the LED blinking) and it worked just fine. This may save you all the trouble 😉 and remove high current/voltages from the board.

    6. If you somehow increase significantly the PWM frequency, make sure that TR2 doesn't get too hot due to the low current drive of R20 which slows the switching time ('Miller effect').

    Good luck.

    (this is an opinion of a non expert so take it with a grain of salt).


  • administrators

    As @zapta says, D5 and D6 protect each other, so no need to change them. If you connect the bed heater directly between +48V and BED- then the voltage across C91 won't change either.



  • @deckingman said in 48v on V_Bed by replacing SMD components?:

    @fcwilt Frederick. If you read what the OP said very carefully, you'll see that he never mentioned driving a bed heater, but rather using the V_bed line to drive some sort of 48V device.

    Hmm... true, very true.

    Still given the other possibilities using the bed heater output and changing parts seems an strange way to go about driving whatever sort of device he is talking about.

    Wonder why he has mentioned what it is? I hope it's not a laser.

    Frederick



  • I would separate the 48V device completely from the Duet to avoid any damage. If it's a laser, one has to control the current with a separate laser driver anyway to limit the current to avoid damaging the laser. To turn on/off or other control, I would use one of the communication protocols and lines to the Duet. To be sure, an optocoupler can be used.


 

Looks like your connection to Duet3D was lost, please wait while we try to reconnect.