TL072: TL072CDR

Hi Rajani,

Overall there are some differences but most performance is better.  It is likely that what you are seeing is a result of the device startup or being measured outside of the recommended operating condition.  The PCN lists some changes and more PCN information can be requested by emailing this PCN support list: "PCN_ww_admin_team@list.ti.com".

1. VOL, VOH improved (rail to rail output) at open load and most other loads (less than 10mA).
2. Output current limit is active instead of passive. Output drops suddenly during current limit. Previous was smooth voltage dropping verses all current loads, high output resistance.
3. Phase reversal was removed for IN+ voltage near V-. Input common mode range increased. Output is correct phase when at least one input in the valid common mode range.
4. Input to V+ diodes added. Input current flows for IN > V+.
5. Typical voltage noise increased 37nV/rtHz versus 18nV/rtHz. Less variance in 1/F noise.
6. Gain band width product increased 5.25MHz versus 3MHz

All of the above differences are compliant with the data sheet specifications and do not represent a quality issue. 

If you send a schematic and a description of the problem, we can see if we can identify where the problem may be.

Regards,
Mike

Hi,

But in that PCN I did not found information such that any electrical changes, Is there any other PCN has released for this part mentioning that above spec changes in datasheet, could you share that PCN no. details.

Below are the expected Input & output expected results along with application circuit.

Basically Ultrasonic return echo signal processing purpose.

Thanks,

Rajani.M

Output of final stage of above is again feeded to one of op amp (TL062C) comparator input +Vin.

Could you plz simulate above circuit operation in simulation & confirm its working, suitable remedial to resolve.

I already mentioned production stopped.

2500 Qty parts are available with us.

Thanks in advance,

Rajani.M

Raj,

The signal you are showing has the input centered around GND; is that correct?  With a gain of ~5.6, and the input at GND, and the non-inverting terminal (pin 5) at 5.5 V; this would cause the output to have so much gain the output would be in the rail.

Can you verify the input signal?

What is the actual failure?  Do you have data on the failed device?

There is no additional PCN; more information can be requested about this PCN by writing to the PCN team:

PCN_ww_admin_team@list.ti.com

 Regards,
Mike

Hi,

Input signal on Pin 6 centered on 5.5V DC, its not Ground reference, Inverting pin input

At Non Inverting pin 5, directly DC 5.5V is given

So, I think common mode voltage between both is 5.5V , only the difference between them is + / -1 V max  will amplify, So output is hardly 5.6 V at pin 7.

that to diodes are there in feedback.

Max 10.3V is the pin1 output is expected. Still 0.7V buffer is there to output swing.

whether new PCN changed part can support till output of 10.3V DC with 11V supply.

Is it Rail to rail Input & output type. Also confirm

Thanks,

Rajani.M 

Hi Raj,

Ok.  So if there is no AC signal, then the input and output settle around 5.5V DC?  That makes more sense.

Yes, the output should actually be improved for rail-to-rail performance - I don't think this would be a problem.  The input is NOT rail-to-rail, but is no different than the previous version, which appears to be OK on this design.

One difference that has come up a few times is the input diode; on the old silicon, if the voltage at the input pins exceeds the supply pin, no current will conduct through the input.  But on the new silicon, there is a diode from the inputs to the supply, so current will conduct.  If there is a case where the input can be higher than the supply, it may cause different measurement results.

I still haven't seen any information on what the actual problem is; do you have any waveforms or data on the difference between the two?

Regards,
Mike

Hi Michael,

  Thanks for your response, if possible could you plz provide me TI simulated file with above our application circuit.

  Input on Inverting pin via series resistor is 10K 0n 5.5V DC apply 2V max pk -pk waveform as shown in above figure wise.

  I have ok device waveforms waiting for faulty device waveforms stage by stage (Actually team is on Christmas holidays) 

  Currently I am analyzing theoretically is this works for our application requirement.

  Is there any drop in replacement device of TL07CDR device for cross reference.

Thanks,

Rajani.M

Hi Art Kay,

With this simulated file, you are confirming that the input/output voltage levels of the op-amp do not exceed the supply rail with a +/- 500 mV. input, and it functions well.

Could you please change the input to ±1.0V, ensure it meets the device specifications, simulate it. and share the results with me along with expression (formula) of output at each stage.

I guess it will exceed the supply limit range with 1V input on 5.5V DC source.

Thanks,

Rajani.M

Hi Kay,

Could you plz include below schematic (subsequent circuit also added) in simulation & share the results with above mentioned Input.(on 5.5 V DC superimpose +/-1V peak)

Thanks for the details in advance.

Thanking you,

Rajani.M

Hi Kay,

Thanks for your response.

I am actually looking for a simulation using the exact device TL072CDR.

For the last 20 years, this device has worked well. However, recently we procured material from a factory in Malaysia, following a PCN (Product Change Notification), and observed an issue. Previously, we used to procure from Mexico."

I understand your concern. If the design had a loading effect issue, it indeed seems unlikely that it would have worked well for the past 20 years. The recent issues you're experiencing might be due to differences in the manufacturing process or materials used by the new supplier in Malaysia compared to the previous supplier in Mexico.

Thanks,

Rajani.M

Hi Team,

I have ordered samples on TI website for this part TL072CDR part & received sample have batch code details of 3CK, AE7164.

with this functionally working fine, attaching here pic. for your reference

where as currently I am facing the problem with 48K, AKKR 64 batch which are there production lot 2500 no.s...attached pic is for your reference

I guess is there any specific batch issue / Manufacturing site location shall we decode from top marking.

batch code details decoding plz suggest also.

Thanks,

Rajani.M

Hi Raj,

The first two number/letter combinations of the middle marking indicate the date the device was manufactured.  For the first device, the middle term is "3CK", this means it was manufactured in 2023 and the "C" refers to December (dates are: Jan = 1, Feb = 2...Dec = C).  The second device was manufactured in 2024, in August (8).  However, the PCN states that devices could have been shipped as early as  September, 2022.  The only way to know from the marking if it is old/new die is if it pre-dates Sept. 2022.  Since that time, it is possible that it could be either die.  Unfortunately we have no way on the apps. support team to give you any further insight to specific lots.

I hope this helps,
Mike

Hi Michael,

I am also trying to provide good & defective IC waveforms, it's my bad even recently received samples also with old die, so it worked well.

Defective ICs are stucked at manufacturing location, from your end Is it possible to provide me IC with new die manufactured, so that I can test & share with you waveforms.

By comparing old & new data sheet I also found functional block diagram change, older one with transistor based, recent one with FET based.

Above I have shared my circuit design details with input range, did you able to see any flaw with it. Input is on 5.5 +/ 1V range.

Is there any input & output impedance mismatch due to this after PCN, datasheet parameters wise no change but.

Would you suggest any immediate alternative for this part, so that production stoppage can be cleared, until our analysis completion.

Thanks for your support in advance.

Thanks, 

Rajani.M

Raj,

The old device / data sheet had JFET inputs and the new device has MOSFET inputs.  Both JFET and MOSFET should have very high impedance.  One important advantage of JFET over MOSFET is better 1/f noise.  One important advantage of MOSFET over JFET is input common mode range limitations.  The switch from JFET to MOSFET may cause an increase in 1/f noise.  Normally increased noise would only degrade performance, but in your case I think you have a gross functional issue so I don't think noise is the issue.  The switch from JFET to MOSFET should improve common mode range, but this should not negatively impact your application.  In general, for device changes the specification table should change if the maximum limit becomes worse, but does not need to change if the performance becomes better.  In your case, it may be that what is normally considered an improved specification has a negative impact on your circuits operation, but we cannot speculate on what that would be without some details on actual circuit performance issues.

Best regards, Art

Hi Art Kay,

Thanks for bearing with me, I am actually received defective IC part (New Die device) Yesterday, Captured all required waveforms along with schematic are attached here, In this new die device outputs are not as per datasheet.

LIDEC Test results for TI.pptx

Looking for your valuable advice. to resolve this issue.

Thanks,

Rajani.M

Thanks for your reply....

Actually we are not using H sufficx part, our Actual part is TL072CDR.

Option1 : I will try with C9 place fast switching diode placement (If possible plz share me Pspice simulated file with diode inclusion) 

Option 2: trying to reduce the gain to limit Output.

One more observation supply current consumption also changed with new die part.

Thanks,

Rajani

Hi,

Plz update datasheet parameter forTL072CDR Min peak output swing parameter w. r. to recent PCN wise.

Thanks,

Rajani.M