6
\$\begingroup\$

Consider the following circuit:

schematic

simulate this circuit – Schematic created using CircuitLab

It is used to supply a 6VDC device (represented by R3) with a 12V car battery. More context on why this opamp solution was chosen instead of a much simpler solution like using an LDO, is discussed in this question. Basically the point is that Vin must follow Vout so the device can get an indication if the battery is running low.

Based on the circuitlab simulation, it will not work with an OP295: Vout (VM1) is 3.4V instead of 6V. So it is time to find a suitable component.

The device uses 50mA when active, 10uA when in sleep mode. An opamp that seems to work is a ST TSX631 (based on LTspice simulation, see below) but it`s only available in a SMD package and I am specifically looking for a THT component. I could of course go and get the tools to work with SMD components, but the point of this question is much more about learning why a TSX631 works in this simulation, and a different one (in this case an OPA251) doesn't.

Simulation of the opamp circuit using a TSX631 in LTspice looks good (using this model from ST.com website), Vout is 6V:

enter image description here

However, when trying an OPA251 (which should be able to handle the load), the output voltage doesn't go up to 6V and Iout is far too low:

enter image description here

Note that with the TSX631 simulation, LTspice shows a warning:

LTspice 24.0.12 for Windows
Circuit: * D:\Downloads\Draft3.asc
Start Time: Mon Jun 10 11:15:55 2024
solver = Normal
Maximum thread count: 7
tnom = 27
temp = 27
method = modified trap
WARNING: Less than two connections to node u1:vb_2_vref.  This node is used by e:u1:58.
Instance "m:u1:_nmos2": Length shorter than recommended for a level 1 MOSFET.
Instance "m:u1:_nmos2": Width narrower than recommended for a level 1 MOSFET.
Instance "m:u1:_nmos1": Length shorter than recommended for a level 1 MOSFET.
Instance "m:u1:_nmos1": Width narrower than recommended for a level 1 MOSFET.
u1:diode_novd: Emission coefficient, N=0.001, too small, limited to 0.1
Direct Newton iteration for .op point succeeded.
Heightened Def Con from 5e-06 +++++++++++++++++++++++++++++to 5.00112e-06
Heightened Def Con from 5.00112e-06 +++++++++++++++to 5.00225e-06
Total elapsed time: 0.236 seconds.

When selecting an alternative, I am filtering for opamps that can handle the load (Iout >= 50mA, Vout > 12V), with low supply current (it's a battery powered device). But it can't find one that gets results similar to the TSX631.

What steps do I need to take to find a THT alternative for the TSX631, which can be simulated succesfully with LTspice?

\$\endgroup\$
6
  • 1
    \$\begingroup\$ Why do you assume OPA251 should be able to handle the load? 50mA is the max short circuit current at a limited temperature range for +/-15V \$\endgroup\$
    – MrGerber
    Commented Jun 10 at 12:35
  • 1
    \$\begingroup\$ OPA251 does not have nearly enough drive (catalog value 21mA), so it's not strange that that does not work. TSX631 has Iout=51mA for Vout=Vcc when Vcc = 10V and Iout=92mA for Vout=Vcc when Vcc=16V, so you'd expect that to have more than enough drive. \$\endgroup\$
    – MrGerber
    Commented Jun 10 at 12:38
  • 1
    \$\begingroup\$ As mentioned by @MrGerber I've clearly misread the OPA251 datasheet. It's not able to handle the required output current. Testing with another alternative (TLE2062) did yield the correct result. And thanks for the insightful post by simon-fitch, his alternative solution using a transistor makes it a lot easier to find components. \$\endgroup\$
    – Juicer
    Commented Jun 10 at 13:05
  • \$\begingroup\$ Simon's approach would work, but you might get better answerss if you supplied more detail about what you're trying to do. To start with, why can't you measure the battery, itself? What's doing the measuring? Also, if you're concerned about the load presented by your circuit, and the 12V supply really is a car battery, you shouldn't be. \$\endgroup\$
    – stretch
    Commented Jun 10 at 22:56
  • 1
    \$\begingroup\$ @stretch the point of this question is much more about learning why a TSX631 works in this simulation, and a different one (in this case an OPA251) doesn't. By learning that, I'd had hoped to get a better understanding of the relevant parameters in the datasheets. \$\endgroup\$
    – Juicer
    Commented Jun 11 at 5:15

1 Answer 1

10
\$\begingroup\$

Current boost using a transistor:

schematic

simulate this circuit – Schematic created using CircuitLab

Just like a regular follower:

$$ V_{OUT} = V_{REF} $$

You can use a dirt cheap op-amp, like the old-but-good LM358, instead of having to find an expensive power device, and almost any cheap NPN BJT. Using the 2N3904, you can draw up to 100mA through the load without any heatsinking.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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