This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

HDC3021: Out of Calibration, start from our stock,

Part Number: HDC3021

Tool/software:

Products using HDC3021 (this variant has the Kapton Protection Cover) are out of calibration, and the normal calibration adjustment does not allow enough correction.

I have tested 2 product ranges. Both use the HDC3021 on different PCBs, which are manufactured by different subcontractors. They are also different batches of sensors.

Both do not remove the Kapton tape, follow the datasheet for holding, and the boards are not cleaned after assembly.

When we received the initial batch of PCBs for one of the assemblies, we calibrated several devices and they were ithin Texas Instruments specifications. 

But now after 6 months in stock, they are all out of spec. The span has dropped, and our device only allow as 5% gain adjustment, and this isn't enough to bring the sensors into specification.

Below is a table of results, where we put 3 new HDC3021 direct from stock with the Kapton tape fitted. Removed the Kapton and started a calibration run. The PCB assemblies are stored with metalized ESD bags, not the pink anti-static polyethylene bags which can out-gas.  

It can be seen that re-conditioning the sensors resolves the problem. But why is there a problem? The units are stocked with the Texas Instruments fitted Kapton Protection!

Reconditioning 1000s of Humidity assemblies is time consuming and unpractical. HELP  PLEASE..

                                 
  As Found Results (New)                            
                                 
  Chilled Mirror HDC3021 (DEV1) HDC3021 (DEV2) HDC3021 (DEV3) AVERAGE   HDC3021 (DEV1) HDC3021 (DEV2) HDC3021 (DEV3) AVERAGE   HDC3021 (DEV1) HDC3021 (DEV2) HDC3021 (DEV3) AVERAGE  
  %RH %RH %RH %RH %RH   Error %RH Error %RH Error %RH Error %RH   Pass/Fail Pass/Fail Pass/Fail Pass/Fail  
  21.1 26.5 25.6 26.1 26.1   5.4 4.5 5.0 5.0   FAIL FAIL FAIL FAIL  
  43.3 45.5 45.0 45.5 45.3   2.2 1.7 2.2 2.0   FAIL PASS FAIL FAIL  
  75.4 72.2 73.1 72.7 72.7   -3.2 -2.3 -2.7 -2.7   FAIL PASS FAIL FAIL  
                                 
  Span %RH Span %RH Span %RH Span %RH Average Span %RH                      
  54.3 45.7 47.5 46.6 46.6                      
  All Humidity measurements taken at 21°C ±0.1°C                            
                                 
  As Found Results (After Re-Conditioning, 5 Hours @ 100°C, 5 Days @ 43%RH (Potassium Carbonate Solution Chamber))            
                                 
  Chilled Mirror HDC3021 (DEV1) HDC3021 (DEV2) HDC3021 (DEV3) AVERAGE   HDC3021 (DEV1) HDC3021 (DEV2) HDC3021 (DEV3) AVERAGE   HDC3021 (DEV1) HDC3021 (DEV2) HDC3021 (DEV3) AVERAGE  
  %RH %RH %RH %RH %RH   Error %RH Error %RH Error %RH Error %RH   Pass/Fail Pass/Fail Pass/Fail Pass/Fail  
  21.1 23.9 23.0 23.5 23.5   2.8 1.9 2.4 2.4   FAIL PASS FAIL FAIL  
  43.1 44.6 43.3 44.2 44.0   1.3 0.0 0.9 0.7   PASS PASS PASS PASS  
  75.4 74.4 74.4 74.8 74.5   -1.0 -1.0 -0.6 -0.9   PASS PASS PASS PASS  
                                 
  Chilled Mirror Span %RH Span %RH Span %RH Span %RH Average Span %RH                      
  54.3 50.5 51.4 51.3 51.1                      
  All Humidity measurements taken at 21°C ±0.1°C                            
                                 
  As Left Result (Adjusted with gain and offset).                          
                                 
  Chilled Mirror HDC3021 (DEV1) HDC3021 (DEV2) HDC3021 (DEV3) AVERAGE   HDC3021 (DEV1) HDC3021 (DEV2) HDC3021 (DEV3) AVERAGE   HDC3021 (DEV1) HDC3021 (DEV2) HDC3021 (DEV3) AVERAGE  
  %RH %RH %RH %RH %RH   Error %RH Error %RH Error %RH Error %RH   Pass/Fail Pass/Fail Pass/Fail Pass/Fail  
  21.1 21.7 21.6 21.3 21.5   0.6 0.5 0.2 0.4   PASS PASS PASS PASS  
  43.1 43.5 42.6 43.1 43.1   0.2 -0.7 -0.2 -0.2   PASS PASS PASS PASS  
  75.4 74.8 74.9 75.2 75.0   -0.6 -0.5 -0.2 -0.4   PASS PASS PASS PASS  
                                 
  Chilled Mirror Span %RH Span %RH Span %RH Span %RH Average Span %RH                      
  54.3 53.1 53.3 53.9 53.4                      
  All Humidity measurements taken at 21°C ±0.1°C                            
                                 
  HDC3021 Allowable Errors  
       
  Cal Point %RH Max Allowable %RH  
  21% 2.0  
  43% 2.0  
  75% 2.5  
  As Per Texas Instruments Data Sheet.  
  • Oh yes, The Michell OptiCal was checked against a calibrated Rotronic HCS2 sensor and differences were < 0.2%. HDC3021s were sent to external calibration house, and they also got similar results to the original as found values, before the reconditioning.

  • Hi Nick,

    Thank you for the detailed measurement report. During the 6 months in stock, how are the HDC3021's stored? What ambient temperatures and humidity are the parts stored at? Are the sensors stored in any pink anti-static material or any LDPE bags? Storing HDC3021 improperly, even with the tape on, can cause the temporary positive RH offsets you are seeing.

    Thank you.

    -Alex Thompson

  • Hi Alex.

    They are stored between 18 - 28 degC and 25 to 60% RH.

    I've attached the datasheet for the bags we use to store them.

    Static Shielding Bag.pdf

  • Hi Alex.

    Also, this is not an offset problem. this is a gain issue. 

    You will see above, the span improves after the re-conditioning of the sensor.

    I've worked with Humidity Sensors since the early 1990s, calibrating with Saturated Salts to using the Automatic OptiCal Calibrators.

    We don't use any of the pink/blue or yellow bags anywhere. and we also don't use the pink bubble wrap on external packing.

    The Kapton protection, what protection does the Kapton tape give. Only against the soldering processes?

    Regards

    Nick

  • Other manufacturers list compatible bags. What do TI use?

    https://stroebel.de/en/bags-pouches/esd-protection/esd-shielding.html

  • Nick,

    The Kapton tape cover will mostly just protect against the soldering process. While it will slow down gasses and humid air from getting to the sensor cavity, those will eventually penetrate (for example, RH response time will go from a few seconds with the tape off to a few hours with the tape on).

    Do you think there is any chance that the HDC3021s were exposed to any potential chemical contaminations during storage and handling? Were the devices tested with the kapton tape on at any of these readpoints? A negative gain shift like this usually indicates some chemical contamination has occured, and baking can often resolve that effect.

    The storage bags you are using look good. 

    Thanks

    -Alex

  • thanks for the reply Alex.

    I agree with you that it looks like some kind of contamination, but I don't know where it came from.

    We currently have two product assemblies in stock. 

    One assembly has around 1600pcs, and another 1000pcs waiting to be shipped to us.

    For the second assembly, we have around 200 pcs in stock.

    Both were built at different times, with subcontractors, and in different parts of the world.

    All our subcontractors use no-clean lead-free solders and do not wash (clean) the boards post-assembly.

    Currently, we are  Re-conditioning small amounts of stock. We don't want to bulk re-condition until we understand what is causing the effect. Otherwise, we might have to repeat it again later.

    I think the response time through the Kapton will be very slow. We normally use two layers of PTFE tape as a filter on some products and this slows the response significantly.

    Is there a test I can perform on the units with the Kapton still fitted?

    Do you want samples to test a TI?

    The only comment I have is, that when we received the items, full calibration was undertaken on a number of devices, and all passes. They have been in our stock for approx 6 months, and now they don't. 

    Below is a product with the HDC3021 which was shipped to a customer in April 2024, the stock had only arrived weeks earlier. The humidity was very accurate without any adjustments.

  • Hi Alex.

    I'm out of the office tomorrow, back on Monday 11th Nov,

  • NIck,

    If you could provide the MSDS (material safety datasheet) for the storage bags that could help indicate a possible contaminant. Is there any chance that during the removal of the Kapton tape for the post 6-months test, there were any chemicals outgassing nearby?

    Please also take some pictures of the sensor cavity under a microscope, to look for any signs of material contamination. 

    -Alex Thompson

  • Hi Alex.

    I have requested the MSDS sheet from the supplier. 

    One of our sub-assemblies has to be soldered, which is for attaching a battery, and the Kapton is removed afterwards.

    The other is a plugin board, and there is no soldering. 

    There is no cleaning of the boards in production. 

    I cannot think of any contaminants.

    yes, I will take photo's of cavity tomorrow.

    Regards

    Nick

  • HDC3021 stock.pdf

    I have taken images of the two different sub-assemblies we use. Both stocked with the Kapton fitted until until we calibrate.

    Still waiting for the supplier to give the MSDS sheet for the bags. 

  • Hi Nick,

    Thanks for looking into potential contamination sources so thoroughly. Were the sensors exposed to light while they were stored, like on shelf perhaps, or even near a window. i wonder if they could have been getting continuous UV light exposure. Or were they kept away from light, like in a cabinet?

    Regards

    -Alex Thompson

  • Hi Alex.

    Our stock room has no windows and only a roller door at one end. The door is closed 99% of the time. and no stock is within 5 metres of the roller door.

    No, we don't use any cabinets. We use stock boxes as shown below or cardboard boxes. But they are always sealed inside the ESD bags

    Still waiting for supplier to give me MSDS sheet for ESD bags.

    How did the sensors look to you?  they looked ok to me, apart from the excess backing for the kapton. The kapton backing is silicone based I guess, so it will not attack the season. The only downside with silicone is it adsorbs humidity.

  • Nick,

    The sensors storage looks good to me, it's not obvious to me where the source of contamination is occurring, but since reconditioning resolves the issue, that may be your only option. 

    Regards

    -Alex Thompson

  • Hi Alex.

    It's not the answer I wanted, but it's not an ideal world. haha

    Previously I reconditioned with the Kapton tape off, is it possible to recondition with it still on?

    Cheers

    Nick

  • Nick,

    Yes you can recondition the sensors with the tape on. The tape will have no effect on the bake. The tape will heavily slow down the RH response time, from seconds to hours, but since the reconditioning lasts for days, it should be fine. I am very sorry that there isn't an obvious cause to the gain shift you are seeing, but I am glad you are able to correct it. Since the HDC3021s pass after assembly, and then have that gain shift in the months afterwards, there has to be some chemical contamination point somewhere in between the two tests.

    Regards

    -Alex Thompson