LM4132: Voltage ref is ~2V instead of 2.5V: LM4132DMF-2.5/NOPB

amin cheraghi

Part Number: LM4132

Hi there,

I have an issue with the 2.5V voltage reference LM4132 and I was wondering if you can help me fixing that.

My output voltageis correct (2.5V) when there is no load but when driving a ~1mA load, it drops to ~2V . Here is my schematics:

When the Cout cap is removed, the Vref goes up to 2.2V. Any idea?

Thanks,

Amin

over 5 years ago

0 kai klaas69 over 5 years ago

Guru 140200 points

Hi Amin,

why is the GND pin of LM4132 not connected to GND?

Kai

0 Marcoo Z over 5 years ago

TI__Mastermind 19370 points

Hi Amin,

I agree with Kai. Can you double check that the ground pin is connected properly.

0 amin cheraghi over 5 years ago in reply to Marcoo Z

Prodigy 200 points

Hi Kai/Marcoo,

Thanks for your reply. This is being designed in a way to be able to adjust the equivalent impedance between Trim+ and Trim- pin of the regulator. To do so, we needed to have a 2.5V with reference to Trim- not the GND. By the way, the Trim- is in the range of 10mV and is noisy. We previously had a Analog Devices part (ADR03BRZ) and it works well. I'll share with you the Trim- waveform.

Thanks,

Amin

0 amin cheraghi over 5 years ago in reply to amin cheraghi

Prodigy 200 points

This is how the Trim- (GND) looks like. Is it correct to say this part cannot tolerate this noise level on the GND? Is there any other part that might have a better ac performance?

0 amin cheraghi over 5 years ago in reply to amin cheraghi

Prodigy 200 points

And here's the output below. As you can see the DC level is ~2.1 rather than 2.5.

0 Marcoo Z over 5 years ago in reply to amin cheraghi

TI__Mastermind 19370 points

Hi Amin,

This is an interesting configuration, very similar to a current source configuration. The device should have no issue with that much noise, at least not enough to drop it 0.5V.
Do you know if a load of 0.5mA will still drop the output voltage?
Can you add a capacitor between Vout and GND?
The only other thing I can think off is to increase Vcc because the device with a load in this configuration requires a higher Vcc.

-Marcoo

0 kai klaas69 over 5 years ago in reply to amin cheraghi

Guru 140200 points

Hi Amin,

if I see this right, then the positive going noise spike at Vtrim- is 1.2V. This will make the input voltage of LM4132 momentarily break down to 2.1V. That's not good. I agree with Marcoo, I would also increase the supply voltage of LM4123 a bit.

Kai

0 amin cheraghi over 5 years ago in reply to Marcoo Z

Prodigy 200 points

Hi Marcoo/Kai,

Thanks for your replies. Since the board is already fabricated, we thought of other solutions and turns out that the other TI part (REF1925AIDDCT) works better for this design. It gives us a stable 2.5V vref.

Thanks,

Amin

Power management

Power management forum

LM4132: Voltage ref is ~2V instead of 2.5V: LM4132DMF-2.5/NOPB