AMC1117 Adjustable Voltage Regulator Question

[daimoku]

I replaced a faulty adjustable 1117 voltage regulator (with a known good regulator) and some defective caps. After replacement, the readings are as follows:

GND Out In
0.51V 3.00V 3.258V
Should be ~ 0V 1.5V 3.3V

Since the output reading is 3.00V instead of 1.5 (and it's getting proper input), does that suggest the SMD cap and/or resistors are fault?

That is, the regulator is not being adjusted properly. We would use V = i * R, right?

The SMD cap CB35 is 0.1uf 16V, R48 is 22 Ohm 1/16 Watt and R46 is 110 Ohm 1/16 Watt. In-circuit, I measure R46 at 25 Ohms, R48 at 20 Ohms.

I will test them out of circuit (R46 especially). I just want to confirm that the SMD caps/resistors are where I need to focus my attention. Thanks in advance for your help!

[Dgtech]

Re: AMC1117 Adjustable Voltage Regulator Question

The voltages sound strange but the one clue is if you are using GND as a reference point to get your voltage readings, you cannot get anything other than 0V if it's the same GND. Sometimes there will be an analog Ground and a digital ground. Make sure you're using the right ground for your reference point then re-measure.

[daimoku]

Re: AMC1117 Adjustable Voltage Regulator Question

Thanks for pointing that out...my fault on that.

New measurements:
GND Vout Vin
0 2.5 2.7

[daimoku]

Re: 1117 Adjustable Voltage Regulator Question

Tested R46 out of circuit and it measures 111 Ohms so it appears to be okay.

What else might cause the output voltage to be incorrect?

[daimoku]

Re: 1117 Adjustable Voltage Regulator Question

I'm going to go back tomorrow and double check the resistors and the Vin from the upstream regulator. It seems like the fault must be in one of those components.

[PowerAmpFreak]

Re: AMC1117 Adjustable Voltage Regulator Question

Please note that the 1117 is available in different variants. There are both fixed and adjustable versions. Check the entire partnumber on original regulator.

[severach]

Re: AMC1117 Adjustable Voltage Regulator Question

An adjustable 3 pin regulator has no ground. What is ground on a fixed regulator is the adjust pin on the adjustable regulators which is where the resistors attach to. If there are resistors on that pin then it's adjustable. At 0.51v the resistors appear to be working. Grounds are readily available on both the input and output caps to the regulator.

There's no need for guesswork for what the output voltage should be once the resistors are known. The input voltage should be whatever supply it draws from. The output voltage can be calculated from the formula in the data sheets. Ignore the term that uses Iadj. You're not designing a circuit. You just want a ballpark figure to see if things are fubar.

Regulators are often put near the devices they power. Get the datasheets on the nearby devices and see what voltages they require.

One thing they do with many regulator variants including the 1117 is to scramble the pins around to make board layout easier. If the pins on the new part are out of order then you won't be doing much regulating which is what it appears you're doing. Regulating from 3.3v to 3.0v seems a bit tight even for an LDO regulator.

If the pin order is unobtainable from the original part because it is broken the order can be deduced fairly easily. The adjust pin has the resistors. The input pin has continuity to some higher voltage power supply. The output pin will be the one that is left and should be unpowered when the part is removed. Don't leave some circuits unpowered too long. Some multi voltage chips don't like being supplied by only some of their voltages.

Listing out all the supply voltages on the board would help you a lot. Whether 3.0v or 3.3v is the supply shouldn't be guessed at. The supply must come from somewhere.

[kikkoman]

Re: AMC1117 Adjustable Voltage Regulator Question

i'd like to add that a fixed voltage variant CAN be adjusted via resistors (the same way as the adj. version) but only in the range up to its default output voltage.
at least most 3-part regulators can (lm317, 78xx/79xx, 1117 etc) .

[PowerAmpFreak]

Re: AMC1117 Adjustable Voltage Regulator Question

The 1117 can work with a dropout as low as 800mV under load. in your case, you measure 300mV drop, that makes me SURE that you've the wrong type of 1117 regulator fitted. I bet you have a fixed output reg instead of the adj version.

[daimoku]

Re: AMC1117 Adjustable Voltage Regulator Question

There's no need for guesswork for what the output voltage should be once the resistors are known. The input voltage should be whatever supply it draws from. The output voltage can be calculated from the formula in the data sheets.

Thanks...I found this in the datasheet, but it doesn't make sense because Vout would always be greater than Vin.
Vout=Vin *(R1 + R2)/R1

Regulators are often put near the devices they power. Get the datasheets on the nearby devices and see what voltages they require.

This regulator helps power a 588 pin ARM MT5372AJ scaler IC. I have been unable to locate a datasheet for this IC.

Listing out all the supply voltages on the board would help you a lot. Whether 3.0v or 3.3v is the supply shouldn't be guessed at. The supply must come from somewhere.

The pinout is GND Vout Vin with the tab facing away and the supply comes from a nearby voltage regulator with 3.3V output.

I was getting these voltages from the service manual, but I found out there are two versions of this board. The version I have uses a different combination of resistors so the voltage is correct. So, I went back to checking for bad caps and I found a couple. Still have to replace them, but hopefully that will take care of it.

i'd like to add that a fixed voltage variant CAN be adjusted via resistors (the same way as the adj. version) but only in the range up to its default output voltage

Good to know!

I bet you have a fixed output reg instead of the adj version

Just cross-referenced the part number with Digikey to verify and I can confirm it's the 1117 adjustable. Thanks for the input.

[PlainBill]

Re: AMC1117 Adjustable Voltage Regulator Question

Thanks...I found this in the datasheet, but it doesn't make sense because Vout would always be greater than Vin.
Vout=Vin *(R1 + R2)/R1

There is a designation problem on this equation, no wonder you are confused!! I suspect they are referring to Vref as Vin. On page 3 of this datasheet they give the correct formula. Vout= Vref *(1+R2/R1) +Iadj*R2. Vref is 1.250 Volts, Iadj=50µA (.00005 A).

Plugging in some numbers, if R1=500ohms and R2=1000 ohms we get Vout=1.250 * (1 + 1000/500) + .05 = 3.800

PlainBill

[PowerAmpFreak]

Re: AMC1117 Adjustable Voltage Regulator Question

There is a designation problem on this equation, no wonder you are confused!! I suspect they are referring to Vref as Vin. On page 3 of this datasheet they give the correct formula. Vout= Vref *(1+R2/R1) +Iadj*R2. Vref is 1.250 Volts, Iadj=50µA (.00005 A).

Plugging in some numbers, if R1=500ohms and R2=1000 ohms we get Vout=1.250 * (1 + 1000/500) + .05 = 3.800

PlainBill

Vref is the internal reference voltage that the regulator is using. The output voltage is calculated by using this reference voltage and the divider.