TB 6600 getting hot

Today was a hot day ambient temperature was about 40 degrees Celsius

And the controlbox was switched on all day.
So when I cut a long cut suddenly one of the drivers got hot and shut down. So the tandem axis pulled skew and messed up the whole thing I was cutting.

My control box has no forced cooling would forced cooling help prevent this problem
Or could it be that the driver is damaged?

What current are they set to?
I have mine at 2A in a shop that gets to 55 degree C in the summer (around 0 right now) and had no issues with them, but i would venture a guess that the one faulting has the chip not laying flat on the cooler, so open it and check if the board/chip are flat and the thermal pad is in place.
Mind you, as a general prevention i did open all of them and also added thermal paste between the chip and pad and heatsink and pad.
All on closed enclosures with no fans.

Yes, poor placemnet and lack of ventilation will casue the  driver to overheat.
Here is a drawing from Lam Technologies
 
Drives at the top of the enclosure with fan above it.

and a real install of the same drivers.
 

A lot of the "TB 6600" drivers actually use a S109AFTG, why the vendors keep advertising the wrong chip and printing false technical data on the devices, I do not really understand. The box says 1/32 microstepping (?), so it is not a TB6600 (which only goes up to 1/16th)


For small motors S109AFTG is O.K.. for bigger motors I would not use them. ( TB6600 I would not use, no matter what motor )
 

As per the data sheet 3A is asking for trouble it is more like 1.5A. The 4A that everybody states (false, as fraud or due to ignorance remains a secret) is the Absolute Max. Rating. (magic smoke boundary?)
 
Quote from data sheet:" The absolute maximum ratings of a semiconductor device are a set of ratings that must not be
exceeded, even for a moment. Do not exceed any of these ratings.
Exceeding the rating (s) may cause device breakdown, damage or deterioration, and may result in
injury by explosion or combustion."

Greetings Timo

ps. my approach would be, a) reduce current to 1.5 A, b) if not working buy other driver.
 

Also, are these really a TB6600?
I bought one of these recently: www.amazon.co.uk/dp/B07SBZ9SM5
I took it apart and found there was lots of space inside (which suited me) but there is no sign of a TB6600 chip in there. I don't know what the actual power IC is, but it was small, square and surface mount.
It turns out that I am not the only one to wonder about this:
www.grbl.org/single-post/tb6600-fake-or-not-but-who-cares
The article above also points out that anything that supports x32 microstepping is _not_ a TB6600.

There is an advantage to this if you don't need more than 3A, there is lots of space inside, and I was able to build a little Hobbing / ELS controller directly into the stepper driver case (3D printing a new lid)

It's TB67S109 IIRC, it is a small QFN packaged chip
my.mouser.com/ProductDetail/Toshiba/TB67...XXc_r7JSUfhOeDjB5zT_
TB6600 is very finicky and dies easily, this S109 is much sturdier and can handle plenty of misuse .