New HDPLEX 200W LPS

[Karin]

On 9/29/2018 at 6:05 PM, oneguy said:

Why is the ripple voltage so high at 1 millivolt when a sigma11 LPS is around 10 microvolts. Is there something I am missing?

Hello oneguy,

 

In general a ripple voltage of +/- 1mV for a (non-audio) linear power supply is not so bad at all. The LT3045 family however is capable of showing ripple and noise figures of < 10uV in real-live design.

 

The relatively "high" ripple of HDPlex is caused by the pre-regulation circuits, market in red. These are switching DC/DC converters and are necessary to achieve an adjustable voltage from 15V to 3.3V @ 2A.

 

In a full blow linear power supply design the LT3045 regulators would need to dissipate about 24W (=6W per LT3045) extra in heat, when the power supply is set from 15V @ 2A to 3.3V @ 2A. Unnecessary to mention that this is from a technical point of view simply not possible without proper heatsink(s).

 

Regards,

Karin

 

   

[Karin]

On 11/18/2018 at 4:55 PM, rickca said:

HDPLEX uses LT3045-1 for the adjustable rail.  The input voltage is always one V higher than the target value.  For example, if the user sets the adjustable rail at 9V, the voltage input for the LT3045-1 is 10V.  So the actual power consumption is always 2W on the circuit not 24W.

 

LT3045-1 has VIOC while LT3045 does not.  The VIOC Pin controls the upstream regulator to minimize power dissipation.  Practically speaking, It is almost impossible to make a effective heatsink for TDP 24W in the tiny surface area of the LT3045 chip (smaller than a finger nail).

 

For more details about the VIOC feature, see this Analog Devices presentation starting at page 26.

https://www.mouser.com/pdfdocs/ADI_LDO_General_Presentation_2018Apr1.pdf

 

Hello rickca,

 

The question of oneguy was:

“Why is the ripple voltage so high at 1 millivolt when a sigma11 LPS is around 10 microvolts. Is there something I am missing?”

 

Our reply was, that we think that the increase of the ripple is caused by the pre-regulation circuits, which you call upstream regulators. As said before these pre-regulation circuits are necessary to achieve an adjustable voltage from 15V to 3.3V @ 2A without having thermal complications.

 

Would you NOT use the pre-regulations circuits and offer an adjustable output voltage from 15V – 3.3V @ 2A, the ripple voltage would probably be lower, but your worst case energy waste would be about 24W extra when the power rail was set from 15V @ 2A to a 3.3V DC output setting running @ 2A. And that is what oneguy perfectly describes as: “Damn engineering trade offs!”

 

Regards,

Karin