TPS61120

Hi,

I need more information to find the root cause.

So you mean when Vbat is less than 4V, you can still get the normal Vout of 5.5V but no LDO output?

Please share the waveform of "LDOEN" Pin, Vout Pin, SW pin and load condition.

BR

Gene

Gene
With less than 4V input, Vout comes up to the assigned 5.5V but the LDO has no output. PGood is low. LDOenable is tied to Vbatt.

See attached schematic as built.

Vbatt has some ripple reflected back from the switcher. Load is less than 100mA

It seems OK with your schematic, except for 10uF input Cin is recommended.
We need to find what really happend on the circuit. So please share your waveform.
BTW, there's two ways help to find the root cause, please give your feedback:
1) if there's no load at output, is the LDO output still 0V?
2)change the LDOIN directly connected to Vbat and LDOEN with a seperated supply, is the LDO output normally?

BR
Gene

I have found a way to minimize current on the LDO at startup and that works fine.

Now, if I have high current peaks on the Vout (5.5V output) the LDO drops out and then comes back when the current is reduced.

Should I be powering the LDO from Vbatt or Vout (boost output)?

Will the converter hit a current limit and shut down?

Hi,
It seems that the load at Vout is heavy and touches the current limit. Powering LDO from Vbat can have a stable LDO output.
We can find that the current limit of the Boost is about 1.3A and LDO output is about 300mA.
Actually, when hitting the current limit, it won't shut down but the Vout drops. This can lead to the drop of LDO.
So please confirm that the Boost current limit is hit with your load condition. If it really happens, it's better to change for a larger current IC.

BR
Gene

Gene
I watch both Vout (5.5V) and Vldo (3.3V) as I increase the load on Vout.

Load on Vldo stays constant at about 110mA.

Vin starts at 4.0V and sags to 3.4V as the load increases.

Vldo will stay active and stable as Vout sags to 5.0V, but it cuts out as soon as I release the load from Vout.

Any clues as to why this happens?

Hi Bruce,

What's your load condition of Vout? It seems that Vout load is heavy and hit the current limit, so the Vout drops.

And I really recommend to post the waveforms so that we can find more information.

BR

Gene

If I use a bench supply (that does not sag) for Vin, the problem is not there. I can bring the load up to 1 amp.
If I use batteries that are low (1.3V per cell x 3 cells), Vldo cut off when I release the load, not when I increase the load.
Under what conditions does the LDO shut down? If there is a small glitch on Vout (which feeds LDOin) how low can that go before the LDO cuts off? It is set for 3.3V and has a load of 110mA

Under what conditions will the LDO shut down completely?

The Vout sags but does not shut down.

The LDO completely drops out.

Please Advise

Please share your waveforms.

It's difficult to find the problem if there's only description of words. LDO should be work normally according to your statement. If Vout glith leads to the problem, then the waveform of Vout when adding and removing the load can help analyze the problem.

BR

Gene

This is Vout (5.5V) during a heavy load event (60Hz audio)

It does fall below 3.3V

The LDO is powered by this 5.5V output.

The LDO drops out completely (zero volts) during this event.

How can I  maintain 3.3V?  I can tolerate sag but not total loss of 3.3V

Hi Bruce,

From the waveform we can find that Vout has already drops below 3.3V, which is the output level of LDO. I think the problem is that the load is too heavy, in this case, LDO won't work normally because LDO's input has already below the target output.

I suggest to connect the LDO directly to the input voltage soure, not to the output of the boost. Or change for another Boost with larger current limit. what's your application and spec of the circuit, so that we can provide you other options.

BR

Gene

OK so I am exceeding the ratings on the boost converter and the output is sagging below the 3.3V (output) level of the LDO.

Some LDO regulators just fall out of regulation when Vin falls below Vout.

Why does this one shut down completely?

Is it just a linear LDO?

Can I just replace it with another LDO that does not drop out?

This problem only occurs when the batteries have lost capacity and the input to the boost droops. 

Might it help to parallel the batteries with a super cap?

Gene
If the LDO is just a linear regulator, would it be better for me to use a LM2623 or some other higher current boost?
The goal is to get 5.5V for a Class D power amp from (3) AA batteries and squeeze as much life from the batteries as possible.
I have other options for LDO regulators that do not drop out below the target voltage.

Hi Bruce,

TPS61230 has a larger current limit but I'm not sure if it is fit for your application. It's better to know the specific load condition.

BR

Gene

The device idles at 110 - 130mA.

When the class D amp is active, we see currents around 600mA with peaks over 1.5A.

If I can just keep my 3.3V stable when the 5.5V boost sags, I am OK.

Why does the TPS61120 LDO shut off completely on low voltage input?

I use a Microchip MCP1703 in the same place and it sags when Vin drops below 3.3V but never drops out.

Please show me the Vin, Vout, SW and LDOEN pin waveform. Word description gives some clues but waveform gives much more.
Without waveform, I can only guess that: since you connected LDOEN pin to Vin (Vbat), if load is too heavy and makes Vin drops a lot, it may disable LDO. Can you see the same performance if using a voltage soure, not a battery?

LDOen is indeed connected to Vin

The problem does not exist with fresh batteries or a stable Vin supply

Vin (battery)

Vboost (5.5V)

SW

Vldo (3.3v)

It's helpful to get the waveforms of Vin.

I think it's the internal logic that turns the LDO off.

LDOEN is connected to Vin while LDOIN is connected to Vout. Vin drops below 2.6V when load is heavy. The internal EN logic of LDO is 0.8*LDOIN, which is below EN threshold. I think it should be the problem. Please connected the LDOEN to a seperated voltage supply as mentioned before or connceted it to LDOIN (Vout) to see if it can hlep.

BR

Gene