0
\$\begingroup\$

My problem is I need to have an amplifier circuit design - schematic + components BOM with specific model information that is able to provide a dimmable output signal (Vout) 0.2-10V <0.8A for a set of 8x LED drivers using Raspberry Pi PWM GPIO pin which is able to provide input signal (VG) of 0-3.3V <16 mA.

My understanding is that this can be achieved using MOSFET amplifier circuit. I hired an LED vendor to design such a circuit and when he tested in the workshop he was unable to lower the Vout to <1 V, and consequently his circuit was unable to switch off the LED lights. The MOSFET he selected was a D882 P33. The LED lights need to have a low signal of 0-0.25A in order to turn OFF. A negative signal will turn the lights back ON again.

I did some workshop testing using a circuit similar to the one below obtained from an online course from ECE UCSB, and reproduced the same problem. I replaced the RBPI using a universal AC adapter 3V 500mA to not damage the RBPI during testing.

During my testing there were four problems I encountered

1) I was also unsuccessful to lower the Vout to <1A using two MOSFET models BF506 and 2N3906. I believe that this requires a MOSFET with RDSON of <0.15Ohms and I couldnt find this value on the datasheet specifications for these models.

2) I was unsuccessful to establish a minimum drain voltage using R1, R2 resistors and was confused by the capacitor in the VG front end of the circuit. When I connected a 100k Ohm resistor for R1 and a tiny resistance <100Ohms for R2 and the VCC connected to the VG front-end circuit, I was unable to control the voltage VG and the voltage would shoot up to >4V. I then removed the R1, R2 and the VCC connections and was then able to control the voltage.

3) I was not able to use the circuit when the capacitor was connected. Again the voltage VG would shoot up to 3.3V and I could not adjust at all, so I removed the capacitor.

4) The behavior of the VG and Vout was inverted from what was expected. When I provided a 0.9V signal VG, I received a 2.05V Vout, and for a 3.3V VG --> 4.32V out. This is opposite to what was expected which is for the higher VG voltage to produce a low output signal Vout.

I'm adding a photo of the LED drivers. The vendor provided two types of models in the system. I realize in the original problem description I described for a single LED driver, but actually the system is 8x drivers connected in parallel. Of the 8x drivers there are at least 2x different models, I have not opened to inspect for the other 6x.

enter image description here enter image description here enter image description here

enter image description here

\$\endgroup\$
  • \$\begingroup\$ which LED dimmerr? \$\endgroup\$ – Jasen Mar 5 at 7:40
  • \$\begingroup\$ Why do you seem to want an analog solution and not a PWM one? \$\endgroup\$ – Chris Stratton Mar 5 at 18:25
  • \$\begingroup\$ (Re:Chris) I am looking for a soution that works with digital PWM GPIO input signal from RBPI, if the way I draw looks like its analog then that would be my mistake. I am learning as I go along. \$\endgroup\$ – Taylor Hickem Mar 6 at 8:38
0
\$\begingroup\$

Rather than explain or fix the circuit you show which does not have a chance of working I will show what will work.

Specs:

0 to 3.3V source ( low Z) must be amplified to 0 to 10 out.
Load Impedance: assumed high Z.

Easiest solution is a rail to rail in & out (RRIO) Op Amp (OA) with a 10V supply. (LDO)

Gain= 10/3.3=3.03 ( round up )

Resistor range: 100k preferred.

schematic

simulate this circuit – Schematic created using CircuitLab

Here is a discrete solution from 0.1V to ~10V

schematic

simulate this circuit

\$\endgroup\$
  • \$\begingroup\$ Thanks for the response! I have a few follow-up questions to understand how to interpret the sketch. Just now I opened in CircuitLab and tried to understand by using the DC Sweep but was unsuccessful, all of the voltages showed negative values. Questions \$\endgroup\$ – Taylor Hickem Mar 5 at 6:08
  • \$\begingroup\$ CLab is lame, I never use it tinyurl.com/y5bwrbah \$\endgroup\$ – Sunnyskyguy EE75 Mar 5 at 6:13
  • \$\begingroup\$ Thanks for the response! I have a few follow-up questions to understand how to interpret the sketch. I haven't yet researched OpAmp circuits so I need some time to study. Just now I opened in CircuitLab and tried to understand by using the DC Sweep but was unsuccessful, all of the voltages showed negative values. Questions 1) could you label Vout 0-10V, and VCC source 12V? 2) What should the voltage be to the input to the JFET? The way I read it the source 0-3.3V is before the R1 100 kOhm resistance and the input to the U1 JFET is some tiny value, is that correct? \$\endgroup\$ – Taylor Hickem Mar 5 at 6:17
  • \$\begingroup\$ no. the circuit you show cannot work as a linear amp. \$\endgroup\$ – Sunnyskyguy EE75 Mar 5 at 6:19
  • \$\begingroup\$ Thanks for the link to the simulator on falstad. I have a few more questions on the differences between the falstad and the Clab circuit. 1) the falstad circuit doesn't show the top part of U2, can I ignore that part? 2) whre is the RBPI signal ? before or after the 100k resistance? If the RBPI signal is the little line that is going from 0-3.3V then what is the 40HZ source supposed to represent? \$\endgroup\$ – Taylor Hickem Mar 5 at 6:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.