Electronics > Metrology

Analog Device article low noise 5V 5 Amp base on LT6658 but ???

(1/3) > >>

Overspeed:
Hello

As I need a reliable PSU in 5 V up to 1 Amp for my current source , I have located this circuit which looks interesting by using only 2 transistors as I don t need 5 Amp , but ?

from my understanding that a voltage regulator more than other thing as I don t see how they filter something , I think the quality of the noise level is the result of the Vin voltage quality ( orange on the circuit )  , which can be a battery pack .

I plan to use battery pack to limit as possible noise and pollution.

Regards
OS

Kleinstein:
I somewhat doubt that the shown circuit could actually deliver up to 5 A. The transistors have a current gain of maybe 50 (could be less at higher current) and thus 100 mA base current needed.  This looks more like an upper limit of 0.5 A.  For more current one would need higher gain transistors  -likely more a darlington configuration.  I am also not sure about the stability against oscillation - it may be limited to a well behaved load.

Andreas:

--- Quote from: Overspeed on January 12, 2025, 12:53:34 pm --- I think the quality of the noise level is the result of the Vin voltage quality ( orange on the circuit )  , which can be a battery pack .

--- End quote ---
Hello,

Noise level is determined by
- low noise voltage reference
- PSRR from VIN to VOUT (so mainly the OP-AMP features)

A darlington will not be possible without violating the VOUT maximum rating of 6V.

It is also strange to use the 267R resistor to divide down the reference voltage to 1 V (with unknown tempco)
and then amplificate back to 5V if you only need a fixed 5V voltage.
The cirquit also has no current limitation.

For 1 A low noise output I would use something like LT3080IT.

With best regards

Andreas

Overspeed:

--- Quote from: Andreas on January 12, 2025, 02:43:04 pm ---
--- Quote from: Overspeed on January 12, 2025, 12:53:34 pm --- I think the quality of the noise level is the result of the Vin voltage quality ( orange on the circuit )  , which can be a battery pack .

--- End quote ---
Hello,

Noise level is determined by
- low noise voltage reference
- PSRR from VIN to VOUT (so mainly the OP-AMP features)

A darlington will not be possible without violating the VOUT maximum rating of 6V.

It is also strange to use the 267R resistor to divide down the reference voltage to 1 V (with unknown tempco)
and then amplificate back to 5V if you only need a fixed 5V voltage.
The cirquit also has no current limitation.

For 1 A low noise output I would use something like LT3080IT.

With best regards

Andreas

--- End quote ---

Hello
Thanks , after Kleinstein assessment , your assessment complete the concerns list.

LT3080 looks easy with 40 uV of noise and if I use then in // as the linked pic I can reduce the noise ( I think ) and perhaps set a filter ? as a RC noise-reduction network ?

Other advantage of the LT3080 that a TO220 component so a robust solution with possible thermal control

TPS7A85 is also a solution but I always doubts regarding the fact to draw current rom a so small IC I link the TI doc , that why I search a more ''old style circuit'' to have no concern with thermal management .

ADM7151 rom Analog looks also interesting but not sure that possible to use them in //.

Regards
OS
 

hli:

--- Quote from: Kleinstein on January 12, 2025, 01:04:57 pm ---I somewhat doubt that the shown circuit could actually deliver up to 5 A. The transistors have a current gain of maybe 50 (could be less at higher current) and thus 100 mA base current needed.

--- End quote ---
The LT6658 can deliver up to 150mA on that output. So it should be fine at 5A
This circuit is linked at the official LT6658 page, and there it also discusses noise performance.

Navigation

[0] Message Index

[#] Next page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod