CD40109B output G high to low transition

Antonio Romano

Other Parts Discussed in Thread: CD40109B

I am using two of the CD40109B to drive the International Rectifier RDHA701FP19A8QK solid state relay. Each of the outputs drives a single input of the RDHA701. Outputs E, F and H work correctly but Output G on both chips does not. On a High to Low transition output G stays high for several seconds before finally dropping to a logic Low. I have checked and the input is definitely at a logic low level. I have replaced the part and have also tried this on several boards with the same result. Swapping to different inputs of the RDHA701 shows that the issue tracks with outputs G and is not related to the inputs of the RDHA701. I am using the CD40109B as a low to high logic level shifter from 3.3V to 5.0V. The RDHA701FP19A8QK is an 8 channel optically isolated solid state relay. The CD40109B drives the gate of a MOSFET via a 1K resistor that is internal to the RDHA701FP19A8QK. The Drain of this Mosfet is connected to a an LED which turns on and energizes the SSR.

I have been able to get output G to switch quicker by adding a 10K pull-down resistor to the output. However, the fall time is on the order of 15 usec which is way out of spec. The other three outputs do not require pull-downs and switch in nanoseconds.rdha701fp10a8ck.pdf

over 8 years ago

0 ShreyasRao over 8 years ago

TI__Guru 55990 points

Hi Antonio ,
This issue is weird that 3 other outputs are fine but not G output . What are the inputs driving the device ?
Please send a schematic and any scope shots if available for the inputs to the CD40109B device .

0 Antonio Romano over 8 years ago in reply to ShreyasRao

Prodigy 140 points

Not really comfortable pasting schematics on a public forum. Can you give me your contact info so that I can email it to you directly.

0 Antonio Romano over 8 years ago in reply to Antonio Romano

Prodigy 140 points

I sent you a private message with the schematic and waveforms.

0 ShreyasRao over 8 years ago in reply to Antonio Romano

TI__Guru 55990 points

Thanks , I got the schematic and plots . will get back to you .

0 ShreyasRao over 8 years ago in reply to ShreyasRao

TI__Guru 55990 points

Hi Antonio,

just to let you know that I placed order to get the parts and will be measuring it in the lab next week .

0 ShreyasRao over 8 years ago in reply to ShreyasRao

TI__Guru 55990 points

Antonio ,

I see it working in our bench setup without issues . My setup :
Vcc=3.3V ,Vdd =5.5V
En=3.3V , inputs =5khz square wave 0V to 3.3V .
outputs are at 4.96V as expected with level translation to Vdd .

0 Antonio Romano over 8 years ago in reply to ShreyasRao

Prodigy 140 points

Are the fall-times of all three outputs the same? Are they within speciation? I also noticed that you are at 5.5V. I am operating at 5V. If things still look good at 5V then I suspect I may have a bad lot of devices.

Tony

0 Antonio Romano over 8 years ago in reply to Antonio Romano

Prodigy 140 points

sorry all four outputs:)

0 ShreyasRao over 8 years ago in reply to Antonio Romano

TI__Guru 55990 points

Antonio ,

I forgot to send you the attachments for the screenshots I had taken while working on it . The Vdd level and Vcc level is 5V and 3.3V .Sorry for the typo mentioning 5.5V.

It seems to me that you could have a bad lot of devices shipped to you . Please check the top side markings and let me know if it actually the same compared against the one from datasheet .

0 Antonio Romano over 8 years ago in reply to ShreyasRao

Prodigy 140 points

These plots are inconclusive. You need to zoom in on the negative edges and compare them. The issue occurs when the device is under a very high impedance load. Placing a probe on the output actually alleviates the issue to some degree as the fall time is greatly reduced. However, what you should see is a huge discrepancy between the fall times of the third output and the remaining outputs.

0 Antonio Romano over 8 years ago in reply to ShreyasRao

Prodigy 140 points

Actually on second look with that time scale you should be seeing the issue unless you have a low impedance load on the output. Did you check all outputs? I'll check the markings and let you know.

Tony

0 Antonio Romano over 8 years ago in reply to Antonio Romano

Prodigy 140 points

Parts are marked CM0109B

0 ShreyasRao over 8 years ago in reply to Antonio Romano

TI__Guru 55990 points

From what you were seeing in your circuit mentioning fall time of "several seconds" , I do not see that happening here with 100uS of resolution. The plots here are for 4 of the output channels and even when you have 10k to ground you saw about 15uS of fall time which is not evident here .
With probe being on 50ohm termination mode the output level was down but didn't see the phenomenon you mentioned .
You can submit the devices to our FA to see what's going on with that if you wish to debug further and believe the device is defective .

0 Antonio Romano over 8 years ago in reply to ShreyasRao

Prodigy 140 points

With the scope on 50 ohm termination mode you will NOT see the issue. Try remeasuring with a high impedance probe.

0 ShreyasRao over 8 years ago in reply to Antonio Romano

TI__Guru 55990 points

The plots shown are measured with 10Mohm termination mode .

0 Antonio Romano over 8 years ago in reply to ShreyasRao

Prodigy 140 points

OK, sounds like the devices then. Are you using the TSOP version of the device?
Is there a lot code on the devices? I could buy some new devices from Digikey but perhaps they are all from the same lot which would not help me. Alternatively, I could unsolder the bad devices and send them in for FA.

0 ShreyasRao over 8 years ago in reply to Antonio Romano

TI__Guru 55990 points

I worked on the TSSOP , CD40109BPWR devices .
I was referring to the top side marking CM0109B to make sure they are genuine TI parts . I have heard about some fake ones too in the market . Not sure about how to genuinely identify them though. If you have ordered them through Digikey they must be good . I am unsure about the packaging and the lot code of the devices .

0 Antonio Romano over 8 years ago in reply to ShreyasRao

Prodigy 140 points

What is the process for returning the parts for failure analysis?

Tony

0 ShreyasRao over 8 years ago in reply to Antonio Romano

TI__Guru 55990 points

please contact ajsimmions at ti.com for FA .

0 Antonio Romano over 8 years ago in reply to ShreyasRao

Prodigy 140 points

Hi Shreyas, any chance you can send me a couple of the devices that you were testing so that I can try them out in my circuit? I'm concerned that perhaps Digikey has a bad lot of these.

Tony

0 Antonio Romano over 8 years ago in reply to ShreyasRao

Prodigy 140 points

Also do you have an email for ajsimmions?

0 ShreyasRao over 8 years ago in reply to Antonio Romano

TI__Guru 55990 points

please contact ajsimmons at ti.com for FA related enquiries .
I won't be able to send the devices to you since it cannot be done once I receive it as per agreement ,sorry .

0 Antonio Romano over 8 years ago in reply to Antonio Romano

Prodigy 140 points

Well this problem just will not go away. I have now built several of these circuits with different sets of the CD40109B and the problem persists. The problem only exists on the third output and requires a weak pull-down to function. My concern is that this issue will persist when I move to the space-qualified versions of this die. I have no issue in using the weak pull-down but in a space application I need to be able to explain this anomaly and ensure that the problem will not arise at end-of-life.

Regards

Tony Romano

0 ShreyasRao over 8 years ago in reply to Antonio Romano

TI__Guru 55990 points

Hi Tony ,

If this will be used as space qualified version , you would need to send the queries to High reliability forum .

They actually space qualify the devices with additional testing on the parts .They would be the best technical guys who could further support on the issue.

Logic

Logic forum

CD40109B output G high to low transition