Comparator soldering station : Controlling the heating element

[Voidugu]

I want to drive a 60 watt 24volt heating element of a 24volt 80 watt transformer. The trigger pin will be connected on a comparator that is capable of sinking up to 18 mA of current. I know that the ideal circuit for the job would be circuit 4 yet the MOC30xx family of triac drivers is incapable of handling 90 mA of current in order to latch the main triac. Moreover, i got 4 optocouplers and again none of them can handle 90mA of current.
> Which circuit do you suggest i use? (i was going to go with circuit 1 to be honest)
> Is there a better way of getting the job done?
> Do you know any other triac drivers capable of handling such currents?

Last question is a bit off topic:
Which would be the best way to switch ac current?
>Triac
> Full wave rectifier and a MOSFET (Tried it and the power wasted in the diodes of the full wave rectifier  is huge)
> 2 MOSFETs connected source to source and the drain terminals connected to the AC(Tried it and destroyed 2 MOSFETs one after the other by unknown cause {note: the MOSFETs were at all times cool, no magic smoke was released in any case} Still haven't figured out why they got destroyed as the circuit was correct at all times. After disconnecting the MOSFETs and measuring them out, some of their pins would be shorted, in one case the gate with drain showing 0 resistance or the drain with source showing 0 resistance. Do you have any idea why this might have happened? {The circuit is shown in one of the pictures. The MOSFETs are IRF630MFP with a max gate voltage of +- 20V})

I am really sorry for the crappy pictures. I know i need to change my mobilephone.

Regards
Void

[megajocke]

Using MOSFET:s like that should work, but with the schematic you showed it is required that the 24 V supplies for heater and control circuit be two separate, isolated, supplies.

What triac are you trying to use? Using MOC30xx devices shouldn't be a problem together with typical triacs.

[amyk]

You may wish to look at the schematics of the Hakko 936 and its clones.

[GK]

Has anyone ever considered making a variable v-out (DC) SMPS for one of these low voltage irons? Should give better overall control than this phase control stuff.

[Paul Price]

I have a cheap Hung Chang oscilloscope sitting next to my soldering station and the soldering iron's transformer seriously modulates the display on the scope. MOSly I got used to this, but soon I got FET up, I couldn't stand no more to see this distortion upsetting my beautiful waveforms!

Forget about the "variable out" DC power supply for powering a solder station, I use a PIC processor to provide PWM to control and even made a digital LED temp readout.

I got the DC from salvaged power supplies from discarded Canon Printers, they have small detachable SMPS 30V power supplies modules and two of them, with a small power MOSFET and  PWM, gives me some quiet watts for iron control that the scope can't notice.

What's so cool about junk PC printers is that they are almost every week in the electronic "Free Store" recycling area of my apt complex and they have many salvageable parts and they all use a 30 DC volt approx SMPS power supply to power the stepper motors for paper and printhead positioning. And what is so convenient, some Canon models have easily detachable power modules that are the main actor for the perfect free solution to power your hot rod.

If you're not ready to handle the code to get a MCU to sing your tune, you can just make your own  PWM with a single 555 timer chip with a pot to set 555 sq wave ouput  duty cycle to feed a FET to set the power to set the temperature. No feedback but it works.

Else if you want fancy, add  a lm358 to amplify the thermistor output from the iron and a LM339 comparator to compare the thermistor amplified voltage to a reference voltage. The reference is fed to one side of the comparator through a pot and makes a  temperature adj.  The comparator output turns on/off power using a small N-channel power MOSFET that is in series with the soldering iron heating element  and the printer salvaged 30-volt power supplies. The same Canon supply also is used to power all the other electronics in the control circuit.

And the old power transformer acts like the perfect paper weight to stabilize the soldering handle holder, a place for the sponge and the temperature control and temperature digital readout and even a conveniently placed power switch. The Cannon supplies in their plastic modules sit behind minding their own business.

[Rufus]

Has anyone ever considered making a variable v-out (DC) SMPS for one of these low voltage irons? Should give better overall control than this phase control stuff.

It wouldn't give better control until the thermal time constant of the bit is getting comparable to half a mains cycle.

The attraction would be replacing a heavy/expensive mains transformer with a switcher and having done that variable dc is about as easy as PWM.

There are standards for soldering iron performance. From memory one requires any voltage on the bit to be less than 2mV with respect to the grounding terminal, another I think is 5mV. Switch mode power supply noise is likely to be an issue in meeting that requirement.

[Voidugu]

@ megajocke
Why is there a need for two separate supplies?
The triac i am using is the BTA16 600B
Datasheet: www.haopin.com/PDF/BTA16-600B.pdf

@ Paul Price
The initial idea was to use an ATMEGA328 to do the fancy things i.e. control a triac using zero crossing detection circuitry, have a display to show the set and current temperatures, use a PID algorithm to control the temperature of the iron and have timers to switch off the station when not in use for a set amount of time. After some research i realized that no-matter what technique i used the temperature calculated by the mega and the actual tip temperature would vary by 20 degrees celcius at least. As a result i abandoned the idea and tried to find the simplest temperature controlled solution possible. I tried driving the iron using 24 battery DC voltage, the comparator circuit shown above and an N-channeled mosfet with much success. The tricky part is the AC.

Pplz could you please answer me these two questions?
> Which circuit do you suggest i use?
> Do you know any other triac drivers capable of handling such currents?
 
Thanks again

[Paul Price]

"I tried driving the iron using 24 battery DC voltage, the comparator circuit shown above and an N-channeled mosfet with much success. The tricky part is the AC."

AC, I thought you were using DC???

Did you carefully check with a scope/DVM for ground loops or noise pickup adding noise/offset to your control loop?

Was your PID code oscillating? 

The idea of my using PWM is to have just proportional control, not full PID, once the tip temp error output was near the correct value, a slow integral term was fine to keep the temp to a small delta degree C value, but otherwise from questimating the temperature of the tip from determining the exact point that a small wire sample of solder melts and then my visual estimation of  viscosity of the fluid solder once melted, and experiencing it work great in actual use, I have no way to actually measure the instantaneous soldering iron tip temp when used in situ, say on a 4-layer circuit board, just my MCU's idea of the temperature on my digital LED panel readout based on the thermistor's resistance.

However it works much better than the orig controller built into the soldering station, esp. because of the faster response time. I had (with the orig. 40W AC- triac-transformer circuit) great difficulty to unsolder components from a 4 or 6-layer PC motherboard, it took ages! Not enough power.

Why two supplies, someone asked, ans: to make sure I have enough current when the iron is cold plus a little reserve and they were both free and on hand.

I don't have an accurate schematic on my PC of my breadboarded control circuit tried using a just a 339 comparator and lm358 op-amp by themselves, but it worked well although there was no way to calibrate the temperature adj. so I always had to guess what temperature I was dialing in with the pot.

So I switched over to develop a MCU circuit soon into this endeavor because it gave the use of PWM,  A2D to use as temperature input directly from the thermistor, ability to have accurate thermistor temperature by resistance because of imbedded known thermistor resistance values to interpolate with(from thermistor mfg's accurate resistance v. temp charts), and also the ability to add a digital readout not originally on the cheap soldering station, plus the software approach replaces the comparator and allows me to create a proportional control PWM scheme to reach the temperature and a dithering integral PWM term to control the tip temp around the goal temperature.

As far as the switching regulator's ability to limit the mV's of noise to 2 mV an the tip, I could would not give a solder whisker's damn about that.  The switcher output is very clean, anyway. There is some myth that SMPS power supplies always have to be noisey at their output. The truth is  that it is so easy to filter their output noise which is high frequency stuff, so only small valued choke and a small valued cap at the output creates a very quiet output.

Because of PWM, there are no fast on/off surges of power that can create noise, just a MCU's casual drift towards applying the right power to get the right temperature.

[TerminalJack505]

You can find a lot of TRIAC drivers that are capable of driving >=100ma.

Here's a link to some parts found on Digi-key.  (Hopefully, that works.)

[Paul Price]

In circuits I built to control high AC currents on a washing machine universal motor with a TRIAC, I took about a small cellphone charger apart  and use the board within it to feed into an optocoupler C-E ouput circuit to a resistor to apply a 6V approx - voltage to the gate of the TRIAC. TRIAC's love to be triggered that way, it gave me safe isolation and always 100% reliable triggering.

If you don't need isolation, you can just apply a negative voltage and limited to 30mA, optimal to always trigger the gate of just about any TRIAC.

A positive trigger voltage will almost work as well, but requires a higher current over a reasonable temp range, but about 35-50mA will do this side of Hell, Texas and the north pole.

[GK]

Has anyone ever considered making a variable v-out (DC) SMPS for one of these low voltage irons? Should give better overall control than this phase control stuff.

It wouldn't give better control until the thermal time constant of the bit is getting comparable to half a mains cycle.

The attraction would be replacing a heavy/expensive mains transformer with a switcher and having done that variable dc is about as easy as PWM.

There are standards for soldering iron performance. From memory one requires any voltage on the bit to be less than 2mV with respect to the grounding terminal, another I think is 5mV. Switch mode power supply noise is likely to be an issue in meeting that requirement.

A variable v-out SMPS incorporated inside a linear control loop with the temperature sensing element with properly worked out time constants would be major improvement over a lot of the more primitive phase control schemes. I don’t think that “noise" specification would be particularly difficult to meet and certainly wouldn’t matter for a hobby project. Would be a good project for recycling some of the magnetic components of an old ATX computer power supply.

[Rufus]

A variable v-out SMPS incorporated inside a linear control loop with the temperature sensing element with properly worked out time constants would be major improvement over a lot of the more primitive phase control schemes.

A good control loop will be better than a crap one - no surprises there.

The JBCs have 180W heaters and very low thermal mass bits. A half cycle of mains driving the JBC heater increases the bit temperature by about 0.6C. Any improvement from a DC or high frequency PWM drive will be limited to removing about 0.6C of temperature ripple and speeding up the loop response by on average 1/4 of a mains cycle. Both of which are trivial not major improvements even with the best in class heater power to thermal mass ratio irons.

[GK]

I'm sure a phase control scheme for a low voltage iron could be made to work, for all sakes and purposes, as well as any other scheme.

However replacing the bulky mains transformer and the rest with a direct-off-line SMPS appears to me to be good basis for making something significantly better overall than 90% of the cheap junk on the market.

[Voidugu]

Is it okay and reliable to use a simple pnp transistor to trigger the power triac? or should i go for a triac driver?