variable DC DC converter

Hi all,

I want to make a variable DC DC converter using lm2678. The output voltage and maximum current should be adjusted using digital methods (no pots). I have thought of the following circuit. I cannot yet implement it as the inductor will take some time to arrive.

Will this work? Do you have any other good designs ?

Regards

Problems of the circuit:

- as a trivial point, bootsrap capacitor wrongly connected
- voltage control loop has high gain and can't be stable
- voltage control loop has wrong polarity
- current control loop has wrong polarity
- voltage and current error signals are combined in a way that can't achieve the intended effect

It's not correct to add the error signals, instead they have to be ORed. The setpoint that is exceeded first takes the control. Achieving stable control for both feedback loops isn't easy, particularly with variable load. "Simple switcher" LM2678 has internal loop compensation expecting an external feedback factor <1 for stability.

Hi FvM,
Thank you for you reply !!!!!!!

when you say

FvM said:

Problems of the circuit:
- voltage control loop has wrong polarity
- current control loop has wrong polarity
.

Click to expand...

do you mean the suply voltage connections for the opmps? If so yes that was a mistake of mine...sorry.

I put R1 and R2 as 10k and 3.3k as I need a maximum Vout of about 20V. Is this the gain you are referring to ?

How do I perform the OR function in this circuit ?

This supply is for a DC motor control board. In such a case do you think this circuit (even after the errors you mentioned are corrected) will be stable ?

Before this I had a variable linear regulator but it was getting too hot. Would you recommend the linear regulator and the lm2678 as a pre tracking regulator ?

regards

Wrong polarity means unwanted signal inversion.

In the present circuit, voltage control loop gain is R2/(R2+R1) * -U1A gain * U1B gain. U1A gain is OP open loop gain, U1B gain 2

For the OR function, a diode circuit may be used, or a different structure where e.g. the current control cuts back the voltage setpoint above the current limit.

dmta said:

I want to make a variable DC DC converter using lm2678. The output voltage and maximum current should be adjusted using digital methods (no pots).

Click to expand...

Hi [dmta],

I have need for the same thing, except using the MP2307 converter. It's similar enough in operation to the LM2678 that the supporting circuitry can be nearly identical, but I have not yet found any preexisting and suitable designs.

Some use digital potentiometers as the voltage divider to generate the feedback voltage, which makes the voltage control part super easy; but a downside is that the resolution is limited (a max of 256 digital taps seems common), so I'd prefer not to use those.

Your circuit combined with [FvM]'s comments have given me some good ideas on how to approach this. I've started on my own design from the ground up, while trying not to copy any aspect of yours exactly.

I hope you will continue to work on your design as well, and post updates as it progresses. In the near future, perhaps we can compare our results and learn something from each other.