Can anyone explain to me why this circuit works exactly as
would be expected at .temp 27 and also .temp 26 but with
.temp 25 or lower, well, see for yourself.
On 11/8/26 20:14, Edward Rawde wrote:
Can anyone explain to me why this circuit works exactly as
would be expected at .temp 27 and also .temp 26 but with
.temp 25 or lower, well, see for yourself.
I wouldn't trust OpAmp models for situations where they are
driven deep into saturation.
Jeroen Belleman
"Jeroen Belleman" <jeroen@nospam.please> wrote in message news:11cqjgq$knqj$1@dont-email.me...
On 11/8/26 20:14, Edward Rawde wrote:
Can anyone explain to me why this circuit works exactly as
would be expected at .temp 27 and also .temp 26 but with
.temp 25 or lower, well, see for yourself.
I wouldn't trust OpAmp models for situations where they are
driven deep into saturation.
Jeroen Belleman
But this does exactly the same thing.
.temp 27 is fine but .temp 25 and what the??
On 11/8/26 21:06, Edward Rawde wrote:
"Jeroen Belleman" <jeroen@nospam.please> wrote in message news:11cqjgq$knqj$1@dont-email.me...
On 11/8/26 20:14, Edward Rawde wrote:
Can anyone explain to me why this circuit works exactly as
would be expected at .temp 27 and also .temp 26 but with
.temp 25 or lower, well, see for yourself.
I wouldn't trust OpAmp models for situations where they are
driven deep into saturation.
Jeroen Belleman
But this does exactly the same thing.
.temp 27 is fine but .temp 25 and what the??
Hmmm. I'm still running V4.21s and both versions simulate
just fine, even with .temp 25 or lower.
In fact, it still works the same with .temp -270. Clearly
temperature is ignored in the model of the LT1679 I have.
I suppose what I have here can't be of further use then.
Jeroen Belleman
On 11/8/26 21:06, Edward Rawde wrote:
"Jeroen Belleman" <jeroen@nospam.please> wrote in message news:11cqjgq$knqj$1@dont-email.me...
On 11/8/26 20:14, Edward Rawde wrote:
Can anyone explain to me why this circuit works exactly as
would be expected at .temp 27 and also .temp 26 but with
.temp 25 or lower, well, see for yourself.
I wouldn't trust OpAmp models for situations where they are
driven deep into saturation.
Jeroen Belleman
But this does exactly the same thing.
.temp 27 is fine but .temp 25 and what the??
Hmmm. I'm still running V4.21s and both versions simulate
just fine, even with .temp 25 or lower.
In fact, it still works the same with .temp -270. Clearly
temperature is ignored in the model of the LT1679 I have.
I suppose what I have here can't be of further use then.
Jeroen Belleman
Because the LT1010 macromodel is natively built into LTspice, its raw SPICE syntax is encrypted or compiled within the software's monolithic standard library file (LTC.lib). Analog Devices does not distribute the standalone unencrypted .SUBCKT text code for this specific, legacy model on their website.
Most commercial operational amplifier (op-amp) SPICE models do not accurately account for temperature because they are heavily simplified macromodels.
Why Most Op-Amp Models Ignore Temperature
rCo Macromodeling Abstraction: Rather than modeling the dozens of individual internal transistors, diodes, and resistors at the silicon level, standard vendor op-amp models use behavioral voltage/current sources and linear equations.
rCo Fixed Parameters: Internal parameters like open-loop gain (\(A_{OL}\)), input offset voltage (\(V_{OS}\)), input bias current (\(I_{B}\)), and slew rate are often hardcoded to typical room-temperature values (25-#C or 27-#C). Changing the global simulator temperature via .temp or .step temp in a tool like LTspice Temperature Management will frequently yield little-to-no change in the op-amp's internal electrical behavior.
rCo Missing Die-Level Physics: True temperature effects (such as thermal drift of \(V_{OS}\), bandgap voltage shifts, or changing bias currents) require full transistor-level (transistor-level subcircuits) or specialized electrothermal models, which are rarely provided by manufacturers for general-purpose op-amps.
The best you can do:
How to Verify or Work Around It
rCo Check the Datasheet vs. Simulation: Always overlay manufacturer datasheet graphs (such as \(V_{OS}\) or \(I_{B}\) vs. free-air temperature \(T_{A}\)) against a .step temp simulation of the specific part (e.g., Texas Instruments THS4001 SPICE Model Performance) to see if the vendor actually embedded temperature dependence.
rCo Inject Temperature Drift Manually: If your op-amp model is static, you can manually parameterize temperature-sensitive error sources (such as placing a small voltage source in series with an input to mimic temperature-dependent input offset drift, or using .param and behavioral sources).
Now you know.
Can anyone explain to me why this circuit works exactly as
would be expected at .temp 27 and also .temp 26 but with
.temp 25 or lower, well, see for yourself.
One practical reason for the simplified macromodels is that a much more detailed device-level model would take too long to run on SPICE,
especially a transient sim.
| Sysop: | Amessyroom |
|---|---|
| Location: | Fayetteville, NC |
| Users: | 76 |
| Nodes: | 6 (0 / 6) |
| Uptime: | 56:24:23 |
| Calls: | 1,265 |
| Calls today: | 1 |
| Files: | 1,362 |
| D/L today: |
3 files (2,684K bytes) |
| Messages: | 285,806 |