Author Topic: What is the resistance of the Hakko FX-888D thermistor/heater at higher temps?  (Read 2561 times)

0 Members and 1 Guest are viewing this topic.

Offline WarFreak131Topic starter

  • Contributor
  • Posts: 40
  • Country: us
I'm making my own PID controller for a soldering iron using an Arduino, and as a first step, I'm going to try to control the iron that comes with the Hakko FX-888D station.  I was able to get a cheap knockoff for $11.

The manual says that the thermistor resistance at room temperature will be between 43 and 58 Ohms, but there's no mention of what the resistance is at higher temperatures.  I am going to hook it up with another resistor in series and hook it up to 5V and then measure the voltage across the therimstor.  Since the Arduino can only read a max of 5V with a 5mV resolution, I need to know what other resistor to choose to get me a nice range of voltage values across the normal temperature range of 300C to 450C.

With regards to the heating element, does it exhibit the same rise in resistance?  Typically how much current is passed through the element?

Thank you in advance.
 

Offline thm_w

  • Super Contributor
  • ***
  • Posts: 7527
  • Country: ca
  • Non-expert
Profile -> Modify profile -> Look and Layout ->  Don't show users' signatures
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7681
  • Country: ca
The FX-888 station uses A1560 heating element which has a separate temperature sensor, a wire-wound RTD inside. Around 43-58R at room temp but not platinum wire, but nickel-iron I think. Use the math from a ~50R RTD alpha 0.00527 and you can make a resistance table matching the thread below. I think it's the same sensor as the 936:
https://www.eevblog.com/forum/reviews/genuine-vs-fake-hakko-936-ceramic-heater-a1321/25/

A heating element can also act as a temperature sensor - its resistance increasing with temperature.
Hakko used this property for their 470-472 desoldering stations but later gave up and went to a dedicated thermocouple instead. I think it (NiCr heater wire) was too non-linear and without a MCU to correct, the analog controller could not do it accuratge enough.
A1560 is rated 26V 65W 3 ohms. Doesn't jive, I doubt it gets 8.6A.
 

Offline WarFreak131Topic starter

  • Contributor
  • Posts: 40
  • Country: us
Should give some ideas:
https://arduinoplusplus.wordpress.com/2016/08/15/diy-soldering-station-with-hakko-fx-888-iron/
http://dangerousprototypes.com/forum/index.php?topic=1921.0
https://www.eevblog.com/forum/reviews/genuine-vs-fake-hakko-936-ceramic-heater-a1321/

You may consider a T12 clone as well.

Thank you, I will review these.  For now, I want to stick with non-thermocouple variants.  That requires a whole separate implementation of in amps that I'm just not ready for yet.

The FX-888 station uses A1560 heating element which has a separate temperature sensor, a wire-wound RTD inside. Around 43-58R at room temp but not platinum wire, but nickel-iron I think. Use the math from a ~50R RTD alpha 0.00527 and you can make a resistance table matching the thread below. I think it's the same sensor as the 936:
https://www.eevblog.com/forum/reviews/genuine-vs-fake-hakko-936-ceramic-heater-a1321/25/

A heating element can also act as a temperature sensor - its resistance increasing with temperature.
Hakko used this property for their 470-472 desoldering stations but later gave up and went to a dedicated thermocouple instead. I think it (NiCr heater wire) was too non-linear and without a MCU to correct, the analog controller could not do it accuratge enough.
A1560 is rated 26V 65W 3 ohms. Doesn't jive, I doubt it gets 8.6A.

I was planning on using a SEPIC/buck/boost to get the right output current.  I don't feel comfortable having wall voltages in the circuit, so I was going to try to power this from a 36 W (24 VDC 1.5 A) DC power supply.  But I may have to go back to the drawing board if it takes over 8 amps.  But is it possible to get it up to the right temperature, albeit perhaps with a smaller thermal reservoir?
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7681
  • Country: ca
The Hakko 936/FX-888 heater spec seems weird 2.5-3.5R. I measure 3.3R on my FX-8801 which would only need ~16V for 65W but the power transformer is 24VAC. T12's are typically 8R and at 24VDC that's 72W.

You don't need the textbook full power. People run T12's from laptop bricks at 19VDC (45W) and at 12VDC (18W) it's still usable. But a 12VDC 7Ah gel-cell/car cig lighter with a cheap boost-converter i.e. XL6009 module is also used for a portable solution. KSGER offers a five cell 18650 lithium battery power supply too.

Hakko "Everything you need to know about watts"
"Power doesn't do it. Temperature control does. All you need is enough power to keep the tip hot. Anything more than that is a waste."
 

Online magic

  • Super Contributor
  • ***
  • Posts: 7458
  • Country: pl
I played with a knockoff 936 handle using a knockoff A1321 heater. Its resistance is indeed 2~3Ω when cold, but it increases with temperature and the current drops to something more reasonable. I have never measured it, but I have little reason to doubt that it draws the nominal 2A at 24V, i.e. about 12Ω.

Use a buck controller with cycle by cycle current limit at 1.5A and you are good to go. I used simple hysteretic control and therefore an oversized PSU capable of supplying the inrush current.

Beware that some clone stations use completely different heaters with different characteristics and possibly different temp sensors.
 

Offline WarFreak131Topic starter

  • Contributor
  • Posts: 40
  • Country: us
The Hakko 936/FX-888 heater spec seems weird 2.5-3.5R. I measure 3.3R on my FX-8801 which would only need ~16V for 65W but the power transformer is 24VAC. T12's are typically 8R and at 24VDC that's 72W.

You don't need the textbook full power. People run T12's from laptop bricks at 19VDC (45W) and at 12VDC (18W) it's still usable. But a 12VDC 7Ah gel-cell/car cig lighter with a cheap boost-converter i.e. XL6009 module is also used for a portable solution. KSGER offers a five cell 18650 lithium battery power supply too.

Hakko "Everything you need to know about watts"
"Power doesn't do it. Temperature control does. All you need is enough power to keep the tip hot. Anything more than that is a waste."

Thanks for the tip regarding the T12.  I suppose all I need is enough power to boost the temp in the event that I'm heating a heavy thermal load.  Theoretically, should 24V at 1.5A cause the same temperature rise as 12V at 3A?  Since it's a purely resistive load, all the power would be dissipated as heat, correct?

And that paper on watts was bizarre, not expected from a big company like Hakko.  :D

I played with a knockoff 936 handle using a knockoff A1321 heater. Its resistance is indeed 2~3Ω when cold, but it increases with temperature and the current drops to something more reasonable. I have never measured it, but I have little reason to doubt that it draws the nominal 2A at 24V, i.e. about 12Ω.

Use a buck controller with cycle by cycle current limit at 1.5A and you are good to go. I used simple hysteretic control and therefore an oversized PSU capable of supplying the inrush current.

Beware that some clone stations use completely different heaters with different characteristics and possibly different temp sensors.

My approach was going to be pulse at 1.5A and monitor the resistance.  So assuming that 1A would be a steady state for 350 C, then it would be operating at a 66% duty cycle.
 

Online magic

  • Super Contributor
  • ***
  • Posts: 7458
  • Country: pl
Pulses are not that great because they couple inductively and capacitively into the ground wire and induce voltage spikes at the tip. These may or may not be appreciated by whatever semiconductors you happen to solder. I remember seeing it on my iron, don't remember how bad it was and whether it had enough voltage to exceed maximum ratings of anything but the most finicky, exotic parts. Probably not, but worth checking with a scope. Low frequency AC or DC doesn't have such problem.

Also, you can't pulse 1.5A from a 24V supply into a 12Ω (or less) heater. That's not how the Ohm's law works. The iron will draw between 2A and 8A depending on temperature, and likely shut the PSU down if the pulses aren't short enough to be supplied by capacitors.
 

Offline WarFreak131Topic starter

  • Contributor
  • Posts: 40
  • Country: us
Pulses are not that great because they couple inductively and capacitively into the ground wire and induce voltage spikes at the tip. These may or may not be appreciated by whatever semiconductors you happen to solder. I remember seeing it on my iron, don't remember how bad it was and whether it had enough voltage to exceed maximum ratings of anything but the most finicky, exotic parts. Probably not, but worth checking with a scope. Low frequency AC or DC doesn't have such problem.

Also, you can't pulse 1.5A from a 24V supply into a 12Ω (or less) heater. That's not how the Ohm's law works. The iron will draw between 2A and 8A depending on temperature, and likely shut the PSU down if the pulses aren't short enough to be supplied by capacitors.

I was going to use a buck LED driver to step down the voltage to however much is required to only deliver 1.5A across the heater element.

EDIT: The buck driver I have can be modulated by PWM or by analog.  Would it be much safer to pass the PWM signal through an RC filter and have the driver sample the voltage across the capacitor?
« Last Edit: September 03, 2020, 07:14:10 pm by WarFreak131 »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf