TPSM84209: TPSM84209

user2007042 

Unfortunately, no, the TPSM84209 will not support a VIN to VOUT load for a number of reasons.

1) The TPSM84209 implements a light-load efficiency "Eco-mode" operation that turns off the low-side FET when the inductor current reaches zero.  While this reduces the switching frequency and improves efficiency at light-load for the conventional VOUT to GND load, it will prevent the TPS84209 from sinking current from VOUT.

2) The TPSM84209's soft-start is intended to ramp VOUT up from 0V to the regulated output voltage.  Even if the TPS84209 was able to sink current, this soft-start would create an issue when the output was pulled up to VIN while the converter was off.

While the TPSM84209 does not have an over-voltage protection, OVP would typically be a substantial problem for most BUCK converters and modules in an application like this.  At idle, when the output voltage is off, the converter will have it's output voltage pulled up to VIN, which will typically create an Over Voltage Condition, which could result in the converter turning on it's low-side FET and shorting it's output to ground, potentially damaging the load.

There are a few options however, depending on what your specific needs are.

1) A High-side BUCK using a controller with discrete FETs.  A high-side BUCK "flips a BUCK on it's head"  It is possible to build a non-synchronous version using the TPS40211 BOOST controller and sensing the VIN - VOUT voltage, then level shifting that to ground to feed-back into the TPS40211's input.

If the VIN - VOUT voltage will be fairly high, the efficiency loss of a diode in a non-synchronous high-side BUCK wont be too bad and that could work quite well.

For a Synchronous version you'll need to invert the output of the TPS40211 and feed it into an H-bridge driver to collect the polarity.

2) A standard low-side BUCK that is capable of sinking negative output current with a series switch driven by the Power Good of the converter so that the VIN to VOUT current path is open when the BUCK is not switching and closed only after the BUCK converter has brought VOUT up to the target voltage and completed it's soft-start.  This will also prevent shutting down switching from triggering an OVP that could drive the BUCK converter to short VOUT to GND through the load-side FET.

If you can provide me with some details about the target application, I might be able to help recommend a path to follow.

What is the range of VIN?

What is the range of (VIN - VOUT)?

What is the range of current flowing from VIN to VOUT through the load device?

Is the converter intended to be configured for constant current regulation, constant VIN-VOUT voltage regulation? or Constant VOUT regulation?

Thanks for your reply Peter.

The whole thing is an early idea of driving a number of different laserdiodes that have a common Anode.

The Vin range is 6V - 8V.

The (Vin - Vout) range is approx. 1,5V to 3V

The current range is 1A - 2A

Each laser diode should have its own DC/DC and its own (fixed) current limit and On/Off control.

PCB board space is limited, that is why I was looking for tiny modules.

user2007042 

Thank you, I think I understand.

Typically what I have seen for an application like this is the diodes are connected in series with a MOSEFET which is used to regulate their current and the switcher is used to create a "local ground" and increase efficiency / reduce losses in the current regulating MOSEFET.

Often the MOSFET and sense resistor is incorporated into a single IC along with a current sense amplifier and the amplifier to drive the FET to regulate the diode current.

The switcher creates a fixed floor voltage for each diode, which can be different than the other diodes, in order to reduce the forward drop and thus power loss in, the current regulating MOSEFT.

It sounds like you would like to eliminate the MOSFET and have the switcher regulate the diode current as well as providing the floor voltage.  Unfortunately in that configuration you will not be able to use a standard BUCK converter or module because of the inversion of the control loop.

In a conventional BUCK, Output Voltage, Current, and Feedback are all proportional to each other.  Increasing the current requires increasing the output voltage, which increases the feedback voltage, and the duty cycle is is regulated inversely to the Feedback voltage to maintain a negative feedback loop.

In your application, the lower the output voltage of the switcher, with respect to its ground, the higher the current.

It would be possible to do that with a PWM controller driving a high-side BUCK.

But I am not aware of any modules built in a high-side BUCK configuration.

I can look for some 2A switching BUCKs that are capable of sinking current for the first application, but negative (Sinking) current limits are generally not intended to be precise, they are intended to minimize negative current during a fault condition.

Thank you Peter

Yes you have got it completely.

I understand there is no obvious tiny module for this application.

Instead I will se if I can figure out a discrete solution that is simple (small) enough for the available PCB.

For now thanks for your interest.

Kindly
Per Gaarde-Nissen

Per Gaarde-Nissen,

I am checking with my collegues who support our BOOST products do see if they can recommend an integrated FET BOOST converter that could be used with a high-side BUCK output stage to meet your requirements.  You would need external inductors, capacitors, a current sense resistor and a current shunt monitor, but that is the best I can recommend.