Hi,
The ESP32-H2-MINI-1 datasheet includes a reflow profile presented as a graph with a top temperature of 250C. However it is not very explicit on whether this is a recommended temperature for the recommended profile, or the absolute maximum temperature to use for reflow.
Many of the ESP32 chips, and most of the integrated modules on JLCPCB are listed as "Standard Only" parts, so they end up much more expensive to assemble on a board. Apparently when someone reached out to their support they indicated that the datasheet states 250C, which is above their 255+-5C standard for economic assembly, so all the parts were recently moved to the expensive assembly options.
Is 250C really the absolute maximum ? Would 255+-5C not be acceptable for reflowing ? Due to this issue it is much more expensive to use ESP32 modules in projects (at least when using JLCPCB assembly services), so it would be good to clarify whether this is really a hard requirement.
Maximum reflow temperature for modules ?
Re: Maximum reflow temperature for modules ?
I have also come across this problem today. I am not an expert at this, but here are some additional things I have noticed that might help further discussion.
There is a difference in temperature between the ESP32-S3 chip and ESP32-S3-MINI-1 module. ESP32-S3-MINI-1 requires Standard PCBA, while ESP32-S3 does not. The ESP32-S3 chip has a section 4.7 Reliability in its datasheet that mentions "IR reflow solder: 260 + 0 °C, 20 seconds, three times". However, the ESP32-S3-MINI-1 module datasheet has a peak temperature of "235 - 250°C" listed in section 8.3.1 Reflow Profile. Looking inside the 5 Module Schematics section of the ESP32-S3-MINI-1 datasheet, the only other component that looks like it could be reducing the peak temperature is the crystal. However, reviewing several 40Mhz crystals datasheets on Digikey, they seem to have the same 260°C maximum peak temperature.
Both RP2040 and all (?) ESP-32 chips appear to be fabricated on 40nm TSMC. RP2040 has a Tc of 260.
Something of interest is the Pb-Free Process – Classification Temperatures (Tc) table in the IPC/JEDEC J-STD-020 standard. I don't have a copy of the standard, but it is referenced by others, for instance https://www.ti.com/lit/an/spraby1a/spraby1a.pdf. This document implies that the Peak Reflow Classification (Tc) is based upon volume and height of the chip. For the given chip ESP32-H2, the height is 0.85mm and the volume is 4x4x0.85 = 13.6mm^3. This gives a temperature of Tc of 260°C in the above mentioned table. For ESP32-S3 is 41.65mm^3, which also gives a temperature of Tc of 260°C.
Just for fun - lets take the module and work out the volume and height of it. The height is 2.4mm and the volume is 16.6 x 13.2 x 0.8 + 11.95 x 9.95 x 1.6 = 365.54mm^3. Lookup for these values in the above table gives a Tc of 250°C as its greater than 350mm^3 and 1.6mm – 2.5mm. Was this a factor in determining the 250°C limit in the datasheet?
So a question arises, does J-STD-020 apply to modules or just chips? I might have a look at other modules on Digikey to compare peak temperatures and see if theres any sort of correlation between volume, height and the Peak Temperature given.
There is a difference in temperature between the ESP32-S3 chip and ESP32-S3-MINI-1 module. ESP32-S3-MINI-1 requires Standard PCBA, while ESP32-S3 does not. The ESP32-S3 chip has a section 4.7 Reliability in its datasheet that mentions "IR reflow solder: 260 + 0 °C, 20 seconds, three times". However, the ESP32-S3-MINI-1 module datasheet has a peak temperature of "235 - 250°C" listed in section 8.3.1 Reflow Profile. Looking inside the 5 Module Schematics section of the ESP32-S3-MINI-1 datasheet, the only other component that looks like it could be reducing the peak temperature is the crystal. However, reviewing several 40Mhz crystals datasheets on Digikey, they seem to have the same 260°C maximum peak temperature.
Both RP2040 and all (?) ESP-32 chips appear to be fabricated on 40nm TSMC. RP2040 has a Tc of 260.
Something of interest is the Pb-Free Process – Classification Temperatures (Tc) table in the IPC/JEDEC J-STD-020 standard. I don't have a copy of the standard, but it is referenced by others, for instance https://www.ti.com/lit/an/spraby1a/spraby1a.pdf. This document implies that the Peak Reflow Classification (Tc) is based upon volume and height of the chip. For the given chip ESP32-H2, the height is 0.85mm and the volume is 4x4x0.85 = 13.6mm^3. This gives a temperature of Tc of 260°C in the above mentioned table. For ESP32-S3 is 41.65mm^3, which also gives a temperature of Tc of 260°C.
Just for fun - lets take the module and work out the volume and height of it. The height is 2.4mm and the volume is 16.6 x 13.2 x 0.8 + 11.95 x 9.95 x 1.6 = 365.54mm^3. Lookup for these values in the above table gives a Tc of 250°C as its greater than 350mm^3 and 1.6mm – 2.5mm. Was this a factor in determining the 250°C limit in the datasheet?
So a question arises, does J-STD-020 apply to modules or just chips? I might have a look at other modules on Digikey to compare peak temperatures and see if theres any sort of correlation between volume, height and the Peak Temperature given.
Re: Maximum reflow temperature for modules ?
Looking at other wireless modules on Digikey, its pretty common for modules to have a lower than 260°C temperature:
https://www.ezurio.com/documentation/da ... 340-series (less than 300mm^3; 2mm high = J-STD-020 Tc of 260°C) -> from datasheet: Upper target 250°C, Absolute Peak Temperature 260°C
https://www.st.com/resource/en/datasheet/bluenrg-m0.pdf (approx 240mm^3; 2.3mm high = J-STD-020 Tc of 260°C) -> from datasheet: 240°C Peak temperature; "Soldering phase has to be executed with care: in order to avoid undesired melting phenomenon, particular attention has to be taken on the set up of the peak temperature."
https://mm.digikey.com/Volume0/opasdata ... c-2017.pdf (approx 425mm^3; 1.9mm high = J-STD-020 Tc of 250°C) -> from datasheet: "230°C to 250°C max"
https://content.u-blox.com/sites/defaul ... 033351.pdf (less than 260mm^3; 1.9mm high = J-STD-020 Tc of 260°C) -> from datasheet: "230°C to 250°C max"
https://www.murata.com/products/product ... ype1yn.pdf (39mm^3; 1.1mm high = J-STD-020 Tc of 260°C) -> from datasheet: "250 °C"
Theres many more modules on JLCPCB that are "Standard PCBA only" on closer inspection....
https://www.ezurio.com/documentation/da ... 340-series (less than 300mm^3; 2mm high = J-STD-020 Tc of 260°C) -> from datasheet: Upper target 250°C, Absolute Peak Temperature 260°C
https://www.st.com/resource/en/datasheet/bluenrg-m0.pdf (approx 240mm^3; 2.3mm high = J-STD-020 Tc of 260°C) -> from datasheet: 240°C Peak temperature; "Soldering phase has to be executed with care: in order to avoid undesired melting phenomenon, particular attention has to be taken on the set up of the peak temperature."
https://mm.digikey.com/Volume0/opasdata ... c-2017.pdf (approx 425mm^3; 1.9mm high = J-STD-020 Tc of 250°C) -> from datasheet: "230°C to 250°C max"
https://content.u-blox.com/sites/defaul ... 033351.pdf (less than 260mm^3; 1.9mm high = J-STD-020 Tc of 260°C) -> from datasheet: "230°C to 250°C max"
https://www.murata.com/products/product ... ype1yn.pdf (39mm^3; 1.1mm high = J-STD-020 Tc of 260°C) -> from datasheet: "250 °C"
Theres many more modules on JLCPCB that are "Standard PCBA only" on closer inspection....
Who is online
Users browsing this forum: No registered users and 17 guests