Kit for assembling a soldering station on Hakko T12 tips. Or a soldering iron, which requires a soldering iron to assemble. Once again about the T12 soldering iron Hakko t12 soldering station by hand

What is a sting? Hakko T12? This is a cartridge that includes a soldering iron tip, a heater and a thermocouple. Now they are gaining popularity and the Internet is full of articles about them. Due to the fact that they were repeated by the Chinese, prices for them on Ali are around $4, and on sale you can often buy them individually at a price of around $3. The range of these tips is wide, it is claimed that there are more than 80 models. (By the way, T15 are the same tips, fully compatible with T12)

I was also attracted to these stings after watching the reviews. One of the main points is fast heating. When you are debugging or repairing, you often need to solder one wire or replace some part, and waiting every time for the soldering iron to heat up is annoying, and keeping it on all the time, in addition to reducing the resource, does not make the air in the room cleaner. Here the heating takes place literally in ten seconds, i.e. By the time I dropped some flux and took tweezers, the soldering iron was already ready. It’s also not a bad opportunity to warm up large ranges.

Assemble everything correctly with a purchased soldering iron handle with quick replacement, etc. In terms of money, it’s not very justified, since a ready-made station like BK950D costs $35-40 on AliExpress.

Therefore, I decided to simplify everything as much as possible by refusing to change the tips. In principle, as a rule, only a couple of stings are used, rarely three. I decided to just make a couple of soldering irons to make a two-channel soldering station.

So I bought one T12-KU tip for testing for now.

The tip rod at the end has two contact strips, between them a heater with a resistance of 8 Ohms and a thermocouple are connected in series. Supply voltage up to 24V and current up to 3A. Maximum power is about 70W.

If you look from the far side of the heater, then first there is a plus, then a minus, and the body of the cartridge itself is the ground and serves to ground the tip.

I attached the wires to these belts with a simple twist and crimped them with several heat shrinks.

Two thickenings are visible on the sting shaft. After the second thickening from the tip of the sting, the rod has a low temperature, and here you can already handle it with your hands. At this point I wrapped paper with regular stationery glue.

If you have a ready-made handle for a soldering iron or a suitable tube, then you can already glue in the rod. But since I didn’t have anything on hand, I also glued the pen together from office paper.

Of course, after each layer of paper you need to let the glue dry. After complete drying, I crimped heat shrink on top to make it less dirty and more pleasant to hold.

At the back, to increase rigidity, I filled it with glue (there is literally not a large ring of glue there).

The temperature controller was made analog and was based on a circuit from Chinese regulators. The polarity of the heater is not indicated in the diagram; the plus of the heater is on top of the diagram, the minus is connected to the ground of the circuit.

I just remade it to fit the existing parts. I replaced the 7806 stabilizer with LM317, Q1 2N2222, Q2 AO4407 and added a protective diode D3. I provide a drawing of the printed circuit board, it is made on two-sided PCB, the other side is for an earthen polygon. All SMD resistors and ceramic capacitors are size 0805. Additional shunt capacitors are 0.1 µF, but you don’t have to install them. C4 size B.

The only missing part in this circuit is the P-Mosfet.

I also tried to remake the circuit for N-Mosfet, which are much easier to get or pick out.

WARNING. The circuit does not work when using LM358. I managed to launch it using the TL082 op-amp; he provided his version in the comments.

Zener diode D3 and transistor Q2 took the first ones available. Any zener diode for current >20mA and voltage 6V. A transistor for a voltage of more than 40V and a current of more than 6A (for a power supply of less than 20V, you can install Mosfet from old motherboards, they are usually for a voltage of 30V).

Resistor R15 and voltage source V1, this is the heater and thermocouple of the soldering iron.

So far I have assembled the board according to the Chinese version of the circuit and it looks like this when assembled.

Settings

The circuit requires almost no setup, but you need to connect the heater correctly and adjust the temperature range. Debugging must be carried out with the supply voltage reduced to 9 volts, otherwise, if turned on for a long time at 24V, the tip can become red hot. To determine the correct polarity of the heater connection, I broke the circuit near the variable resistor (I did not solder in the substring resistor) and turned on the regulator. If the soldering iron is turned on with the correct polarity, no power is supplied to it and the LED does not light up. Due to the drift of the op-amp zero, this behavior is possible even with incorrect polarity; to check this situation, warm the tip of the tip for half a second with a lighter. If the polarity is not correct, power will be supplied to the soldering iron continuously.

I had a 10k variable resistor available, so the ratings of the adjustment circuit are slightly different from the original; after adjustment, the adjustment range turned out to be from 260º to 390º. Perhaps I’ll decide to expand the range further by reducing the resistance of the low-resistance resistor R2.

Tests

The soldering iron performed quite well in operation. The heating rate turned out to be really high for about ten seconds (I’ll give you a video).

I didn’t see much of a miracle in terms of power, unless of course you compare it with cheap Chinese stations, which for the most part don’t solder, but just pick their snot. And so it’s quite at the level of simple, but branded stations.

I soldered the adapter with this soldering iron. Although for such a thin sting this is a perversion. Soldering such massive parts cannot be called comfortable; heat transfer is clearly not enough. The video turned out boring and long, so I decided not to post it.

In the end, overall I was quite pleased with the results.

Therefore, I plan to order another sting that is more massive, until I decide which type to choose, type BC or D.

And make a two-channel station from a computer power supply. There are plenty of articles about it; removing 20-24v and 6a from it also doesn’t seem to be a problem. I tried it on, and it seems that after removing unnecessary parts from the power supply board, two regulators will fit into the case. At the same time I'm going to use the unit's fan as an exhaust hood. Now I’m using a 12V fan with a piece of filter from a kitchen hood (the description stated that this felt is like activated carbon), but the thrust of one fan is a little insufficient and I plan to install two.

By the way, here is a view of today's fan that I use as an exhaust hood.

When I get around to doing it, I’ll show you what happened. For now, the soldering iron is simply connected to the laboratory unit. If you power one soldering iron, you can use a power supply, for example, from a laptop; mine from a burnt-out laptop produces 19V and 4.5A, which is quite enough for work.

I also provide a video demonstrating the heating speed of the soldering iron. Of course, for a more massive tip or at a lower supply voltage, the warm-up time may increase.

The list of elements shows the values ​​soldered on the board, the notes indicate the elements on the original circuit.

List of radioelements

Designation	Type	Denomination	Quantity	Note	Shop	My notepad
U1	Operational amplifier	LM358A	1			To notepad
U2	Linear regulator	LM317M	1	LM7806		To notepad
Q1	Bipolar transistor	2N2222A	1	9013		To notepad
Q2	MOSFET transistor	AO4407A	1	IRF9540		To notepad
D1-D3	Rectifier diode	1N4148	3	Diode D3 is missing in the original		To notepad
C2	Capacitor	10 nF	1			To notepad
C3	Capacitor	1 µF	1			To notepad
C4	Capacitor	22 µF	1	1 µF		To notepad
C5	Electrolytic capacitor	470 µF	1			To notepad
R1	Resistor	22 kOhm	1	30 kOhm		To notepad
R2	Resistor	39 Ohm	1	51 Ohm		To notepad
R3	Resistor	100 Ohm	1			To notepad
R4	Resistor	120 kOhm	1	100 kOhm		To notepad
R5, R6, R13	Resistor

Hakko-T12 tips are popular primarily due to the possibility of high-quality temperature control (even in non-original Chinese versions). This is facilitated by the design of the tip, where the heater, tip metal and thermocouple are in direct contact. An important feature of the tips is also the combination of a thermocouple and a heater in one serial circuit - therefore, the driver must alternate heating the tip with measuring the thermocouple voltage. The heater power supply voltage is up to 24V.

There are many driver designs for these tips on the Internet. Of course, implementations based on a microcontroller are of the greatest interest - they allow you to achieve the greatest efficiency of regulation, however, simple analog designs based on op-amps will also work quite well and are very easily repeatable. One of the typical circuits presented below is based on a single LM358 op-amp and uses a powerful P-channel field-effect transistor for control. A 7806 stabilizer is used for power supply and reference voltage.

This design was interesting to me primarily for the practice of mounting in SMD, and in addition, I tried to make a circuit with the ability to be powered by a pulsating DC voltage (transformer + rectifier without smoothing). For this purpose, the circuit contains an decoupling diode D and a relatively capacitive capacitor C5. As for the 7806 stabilizer, it is not advisable to significantly deviate the voltage from 6v, as this will require recalculation of both the reference node R1-R2-VR1 and the constant current mode of transistor T1 (which not only controls, but also limits the voltage at the gate of T2, which important). I didn’t have a 7806, the LM317 was in a TO220 package, but I decided to use a compensation stabilizer on the tl431 (all smd and exactly suitable for the input voltage, which due to the pulsating voltage can exceed 30V on C5).

The only non-smd elements will be transistor T2 (there was no smd) and capacitor C5 (when powered by pure DC voltage, smd ceramics are sufficient). I developed several options for boards (dancing from the option with smd LM317 found on the Internet). With tl431 I got 2 options due to 2 pinout options for tl431 - I ended up with a mirror option. If you repeat, check in advance so that you don’t have to straighten your paws :). The result was a compact 34x23mm board, although the Chinese industrial analogue is still smaller :). The board is single-sided, without jumpers, with a ring ground. The adjustment resistor is external.

All passive elements are 0805, with the exception of R13 - purely because I wanted to lay a wider area under it.

As for the adjustment, you should keep in mind that the tip thermocouple has polarity - it is easy to determine if the tip is heated slightly with a voltage of 9-12V and check the response of the hot tip with a multimeter in millivoltmeter mode. The circuit should work immediately if there are no jambs, but the R1-R2-VR1 chain needs to be selected - with the indicated ratings, a fairly wide control range is obtained - approximately 140 - 480C. You can narrow it by proportionally increasing R1-R2. The board also provides a shunt parallel to VR1, but I didn’t need it.

Links

• Original project http://cxem.net/master/87.php
• A handle of average quality and a pair of T12 tips (the connector on the cable will most likely have to be replaced)

Reading local reviews, I have repeatedly thought about buying a soldering iron with a T12 tip. For a long time I wanted something portable on the one hand, powerful enough on the other hand, and, of course, maintaining the temperature normally.
I have relatively many soldering irons, purchased at different times and for different tasks:
There are very ancient EPSN-40 and “Moskabel” 90W, a slightly newer EMP-100 (hatchet), and a completely new Chinese TLW 500W. The last two retain temperature especially well (even when soldering copper pipes), but soldering microcircuits with them is not very convenient :). An attempt to use the ZD-80 (a pistol with a button) did not work - neither power nor normal temperature maintenance. Other “electronic” little things like Antex cs18/xs25 are only suitable for very small things, and they don’t have built-in adjustments. About 15 years ago I used den-on's ss-8200, but the tips are very tiny, the temperature sensor is far away and the temperature gradient is huge - despite the stated 80W, the tip doesn't even feel like a third.
As a stationary option, I’ve been using Lukey 868 for 10 years now (it’s practically 702, only with a ceramic heater and some other little things). But there is no portability at all; you can’t take it with you in your pocket or small bag.
Because At the time of purchase, I was not yet sure “whether I needed it”, I took the minimal budget option with a K-tip and a handle that was as similar as possible to the usual soldering iron from Lukey. It is possible that for some it does not seem very convenient, but for me it is more important that the handles of both used soldering irons fit familiarly and equally in the hand.
The further review can be roughly divided into two parts - “how to make a device from spare parts” and an attempt to analyze “how this device and the controller firmware work.”
Unfortunately, the seller removed this particular SKU, so I can only provide a link to a snapshot of the product from the order log. However, there are no problems finding a similar product.

Part 1 - design

After a mock-up performance check, the question arose about choosing a design.
There was an almost suitable power supply (24v 65W), almost 1:1 in height with the control board, slightly narrower than it and about 100mm long. Considering that this power supply fed some kind of dead (not through its fault!) connected and not cheap Lucent piece of hardware, and its output rectifier contains two diode assemblies for a total of 40A, I decided that it is not much worse than the one common here Chinese at 6A. At the same time, there will be no lying around.
Testing on a time-tested load equivalent (PEV-100, twisted to approximately 8 ohms)


showed that the power supply practically does not heat up - after 5 minutes of operation, the key transistor, despite its insulated housing, heated up to 40 degrees (a little warm), the diodes are warmer (but do not burn your hand, it is quite comfortable to hold), and the voltage is still 24 volts with in kopecks. The emissions increased to hundreds of millivolts, but for this voltage and this application this is quite normal. Actually, I stopped the experiment because of the load resistor - about 50W was released on its smaller half and the temperature exceeded a hundred.
As a result, the minimum dimensions were determined (power supply + control board), the next stage was the housing.
Since one of the requirements was portability, even the ability to stuff it into pockets, the option of ready-made cases was no longer needed. The available universal plastic cases were not at all suitable in size, the Chinese aluminum cases for T12 for jacket pockets were also too big, and I didn’t want to wait another month. The option with a “printed” case did not work - neither strength nor heat resistance. Having assessed the possibilities and remembering my pioneer youth, I decided to make one from an ancient one-sided foil fiberglass laminate that had been lying around since the times of the USSR. Thick foil (the micrometer on a carefully smoothed piece showed 0.2mm!) still did not allow etching tracks thinner than a millimeter due to side etching, but for the case it was just right.
But laziness, coupled with a reluctance to create dust, categorically did not approve of sawing with a hacksaw or cutter. After assessing the available technological capabilities, I decided to try the option of sawing textolite using an electric tile cutter. As it turned out, it is an extremely convenient option. The disc cuts fiberglass without any effort, the edge is almost perfect (you can’t even compare it with a cutter, hacksaw or jigsaw), the width along the length of the cut is also the same. And, importantly, all the dust remains in the water. It is clear that if you need to saw off one small piece, then it will take too long to unfold the tile cutter. But even this small body required a meter of cutting.
Next, a case with two compartments was soldered - one for the power supply, the second for the control board. Initially, I didn't plan to split. But, as with welding, plates soldered into a corner tend to reduce the angle as they cool, and an additional membrane is very useful.
The front panel is bent from aluminum in the shape of the letter P. There is a thread cut in the upper and lower bends for fixation in the case.
The result was this (I’m still “playing” with the device, so the painting is still very rough, from the remains of an old spray can and without sanding):

The overall dimensions of the case itself are 73 (width) x 120 (length) x 29 (height). The width and height cannot be made smaller, because... The dimensions of the control board are 69 x 25, and finding a shorter power supply is also not easy.
At the back there is a connector for a standard electrical wire and a switch:


Unfortunately, the black microswitch was not in the trash; I will have to order one. On the other hand, white is more noticeable. But I specifically set the connector to standard - this allows, in most cases, not to take an additional wire with you. Unlike the option with a laptop socket.
Bottom view:

The black rubber-like insulator is left over from the original power supply. It is quite thick (a little less than a millimeter), heat-resistant and very difficult to cut (hence the rough cutout for the plastic spacer - it almost didn’t fit). It feels like asbestos impregnated with rubber.
To the left of the power supply is the rectifier radiator, to the right is the key transistor. In the original PSU, the heatsink was a thin strip of aluminum. I decided to “aggravate” it just in case. Both heatsinks are isolated from the electronics, so they can freely adhere to the copper surfaces of the case.
An additional heatsink for the control board is mounted on the membrane; contact with d-pak cases is ensured by a thermal pad. There is not much benefit, but everything is better than air. To prevent a short circuit, I had to slightly bite off the protruding contacts of the “aviation” connector.
For clarity, a soldering iron next to the body:

Result:
1) The soldering iron works approximately as advertised and fits well in jacket pockets.
2) The following items have been disposed of in the old trash and are no longer lying around: a power supply, a piece of fiberglass from 40 years ago, a can of nitro enamel from 1987, a microswitch and a small piece of aluminum.

Of course, from the point of view of economic feasibility, it is much easier to buy a ready-made case. Even though the materials were practically free, “time is money.” It’s just that the task of “doing it cheaper” didn’t appear on my list of tasks at all.

Part 2 - Operational Notes

As you can see, in the first part I did not mention at all how it all works. It seemed to me advisable not to confuse the description of my personal design (rather “collective-farm homemade” in my opinion) and the functioning of the controller, which is identical or similar for many.

As a bit of a preliminary warning, I want to say:
1) Different controllers have slightly different circuitry. Even outwardly identical boards may have slightly different components. Because I only have one specific device of mine, I can in no way guarantee a match with others.
2) The controller firmware that I analyzed is not the only one available. It is common, but you may have different firmware that functions differently.
3) I do not at all lay claim to the laurels of the discoverer. Many points have already been covered previously by other reviewers.
4) Next there will be a lot of boring letters and not a single funny picture. If you are not interested in the internal structure, stop here.

Design overview

Further calculations will be largely related to the controller circuitry. To understand its operation, an exact diagram is not necessary; it is enough to consider the main components:
1) Microcontroller STC15F204EA. An unremarkable chip from the 8051 family, noticeably faster than the original (the original was 35 years ago, yes). Powered by 5V, has on board a 10-bit ADC with a switch, 2x512 bytes nvram, 4KB program memory.
2) A +5V stabilizer, consisting of 7805 and a powerful resistor to reduce heat generation (?) on 7805, with a resistance of 120-330 Ohms (different on different boards). The solution is extremely cost effective and heat efficient.
3) Power transistor STD10PF06 with wiring. Operates in key mode at low frequency. Nothing special, old man.
4) Thermocouple voltage amplifier. The trimmer resistor regulates its gain. It has input protection (from 24V) and is connected to one of the inputs of the MK ADC.
5) Reference voltage source on TL431. Connected to one of the inputs of the MK ADC.
6) Board temperature sensor. Also connected to the ADC.
7) Indicator. Connected to MK, operates in dynamic display mode. I suspect that one of the main consumers is +5V
8) Control knob. Rotation adjusts temperature (and other parameters). The button line in many models is not sealed or cut. If connected, it allows you to configure additional parameters.

As you can easily see, all functioning is determined by the microcontroller. I don’t know why the Chinese are installing just this one, it’s not very cheap (about $1, if you take several pieces) and it’s close in terms of resources. In typical Chinese firmware, literally a dozen bytes of program memory remain free. The firmware itself is written in C or something similar (the obvious tails of the library are visible there).

Controller firmware operation

I don’t have the source code, but IDA is still here :). The mechanism of operation is quite simple.
At initial startup, the firmware:
1) initializes the device
2) loads parameters from nvram
3) Checks whether the button is pressed, if it is pressed, it waits for it to be released and launches the advanced parameters settings subsection (Pxx). There are many parameters, if you don’t understand, then it’s better not to touch them. I can post the layout, but I'm afraid of causing problems.
4) Displays “SEA”, waits and starts the main work cycle

There are several operating modes:
1) Normal, normal temperature maintenance
2) Partial energy saving, temperature 200 degrees
3) Complete shutdown
4) Setting mode P10 (temperature setting step) and P4 (thermocouple op-amp gain)
5) Alternative control mode

After startup, mode 1 works.
When you briefly press the button, you switch to mode 5. There you can turn the knob to the left and go to mode 2 or to the right - increase the temperature by 10 degrees.
A long press switches to mode 4.

In previous reviews there was a lot of debate about how to properly install a vibration sensor. Based on the firmware I have, I can say unequivocally - it makes no difference. Entering partial energy saving mode occurs when there is no changes the state of the vibration sensor, the absence of significant changes in the temperature of the tip and the absence of signals from the handle - all this for 3 minutes. Whether the vibration sensor is closed or open is completely unimportant; the firmware only analyzes changes in state. The second part of the criterion is also interesting - if you solder, then the temperature of the tip will inevitably fluctuate. And if a deviation of more than 5 degrees from the set value is detected, there will be no exit to the energy saving mode.
If the energy saving mode lasts longer than specified, the soldering iron will turn off completely and the indicator will show zeros.
Exit from energy-saving modes - by vibration or by the control knob. There is no return from full to partial energy saving.

The MK is engaged in maintaining the temperature in one of the timer interrupts (there are two of them, the second deals with the display and other things. Why this was done is unclear - the interrupt interval and other settings were chosen the same, it would have been possible to get by with a single interrupt). The control cycle consists of 200 timer interrupts. At the 200th interruption, the heating is necessarily turned off (- as much as 0.5% of power!), a delay is performed, after which the voltages from the thermocouple, temperature sensor and reference voltage from TL431 are measured. Next, all this is converted into temperature using formulas and coefficients (partially specified in nvram).
Here I will allow myself a small digression. Why there is a temperature sensor in this configuration is not entirely clear. If properly organized, it should provide a temperature correction at the cold junction of the thermocouple. But in this design, it measures the temperature of the board, which has nothing to do with the required one. It either needs to be transferred to a pen, as close as possible to the T12 cartridge (and another question is where in the cartridge the cold junction of the thermocouple is located), or thrown away completely. Perhaps I don’t understand something, but it seems that the Chinese developers stupidly ripped off the compensation scheme from some other device, completely not understanding the principles of operation.

After measuring the temperature, the difference between the set temperature and the current temperature is calculated. Depending on whether it is large or small, two formulas work - one is large, with a bunch of coefficients and delta accumulation (those interested can read about the construction of PID controllers), the second is simpler - with large differences, you need to either heat it up as much as possible or turn it off completely (depending on from the sign). The PWM variable can have a value from 0 (disabled) to 200 (fully enabled) - according to the number of interruptions in the control cycle.
When I just turned on the device (and had not yet gotten into the firmware), I was interested in one thing - there was no jitter of ± a degree. Those. The temperature either remains stable or jumps by 5-10 degrees at once. After analyzing the firmware, it turned out that it apparently always trembles. But if the deviation from the set temperature is less than 2 degrees, the firmware shows not the measured temperature, but the set temperature. This is neither good nor bad - the jittery low order is also very annoying - you just need to keep it in mind.

Concluding the conversation about the firmware, I want to note a few more points.
1) I haven’t worked with thermocouples for about 20 years. Maybe during this time they have become more linear;), but before, for somewhat accurate measurements and if possible, a nonlinearity correction function was always introduced - with a formula or table. Here this is not the case at all. Only the zero offset and slope angle can be adjusted. Maybe all cartridges use high-linearity thermocouples. Or the individual scatter in different cartridges is greater than the possible group nonlinearity. I would like to hope for the first option, but experience hints at the second...
2) For a reason unknown to me, inside the firmware the temperature is set as a fixed-point number with a resolution of 0.1 degrees. It is quite obvious that due to the previous comment, 10-bit ADC, incorrect cold end correction, unshielded wire, etc. The real accuracy of measurements will not be even 1 degree. Those. It looks like it was ripped off again from some other device. And the complexity of the calculations has increased slightly (you have to repeatedly divide/multiply 16-bit numbers by ten).
3) The board has Rx/TX/gnd/+5v pads. As far as I understand, the Chinese had special firmware and a special Chinese program that allows you to directly receive data from all three ADC channels and configure PID parameters. But there is none of this in the standard firmware; the pins are intended exclusively for uploading firmware to the controller. The pouring program is available, works through a simple serial port, only TTL levels are needed.
4) The dots on the indicator have their own functionality - the left one indicates mode 5, the middle one indicates the presence of vibration, the right one indicates the type of temperature displayed (set or current).
5) 512 bytes are allocated to record the selected temperature. The entry itself is made correctly - each change is written to the next free cell. As soon as the end is reached, the block is completely erased, and writing is done to the first cell. When turned on, the farthest recorded value is taken. This allows you to increase the resource by a couple of hundred times.
Owner, remember - by rotating the temperature setting knob, you waste the irreplaceable resource of the built-in nvram!
6) For other settings, the second nvram block is used

Everything is with the firmware, if you have any additional questions, ask.

Power

One of the important characteristics of a soldering iron is the maximum heater power. It can be assessed as follows:
1) We have a voltage of 24V
2) We have a T12 tip. The cold resistance of the tip I measured is just over 8 ohms. I got 8.4, but I cannot claim that the measurement error is less than 0.1 Ohm. Let's assume that the real resistance is no less than 8.3 Ohms.
3) Resistance of the STD10PF06 key in the open state (according to the datasheet) - no more than 0.2 Ohm, typical - 0.18
4) Additionally, you need to take into account the resistance of 3 meters of wire (2x1.5) and connector.

The total resistance of the circuit in a cold state is at least 8.7 Ohms, which gives a maximum current of 2.76A. Taking into account the drop on the key, wires and connector, the voltage on the heater itself will be about 23V, which will give a power of about 64 W. Moreover, this is the maximum power in a cold state and without taking into account the duty cycle. But don’t be too upset - 64 W is quite a lot. And given the design of the tip, it is enough for most cases. When checking the performance in constant heating mode, I placed the tip of the tip in a mug of water - the water around the tip was boiling and steaming very vigorously.

But an attempt to save money using a power supply from a laptop has very questionable effectiveness - an apparently insignificant decrease in voltage leads to the loss of a third of the power: instead of 64 W, about 40 W will remain. Is the $6 saving worth it?

If, on the contrary, you try to squeeze the declared 70W out of the soldering iron, there are two ways:
1) Slightly increase the power supply voltage. It is enough to increase it by only 1V.
2) Reduce the circuit resistance.
Almost the only option to slightly reduce the circuit resistance is to replace the key transistor. Unfortunately, almost all p-channel transistors in the package used and for the required voltage (I wouldn’t risk setting it to 30V - the margin would be minimal) have similar Rdson. And that would be doubly wonderful - at the same time, the controller board would heat up less. Now in maximum heating mode, about a watt is released on the key transistor.

Accuracy/stability of temperature maintenance

In addition to power, stability of temperature maintenance is no less important. Moreover, for me personally, stability is even more important than accuracy, because if the value on the indicator can be determined experimentally - I usually do so (and it is not very important that when the setting is 300 degrees, the real value on the tip is 290), then instability cannot be overcome in this way . However, it feels like the temperature stability on the T12 is noticeably better than on the 900 series tips.

What makes sense to change in the controller

1) The controller is heating up. Not fatal, but more than desirable. Moreover, it is mainly not the power part that heats it, but the 5V stabilizer. Measurements showed that the current at 5V is about 30 mA. 19V drop at 30mA gives approximately 0.6W of continuous heating. Of this, about 0.1 W is released at the resistor (120 Ohm) and another 0.5 W is released at the stabilizer itself. The consumption of the rest of the circuit can be ignored - only 0.15 W, of which a noticeable part is spent on the indicator. But the board is small and there is simply nowhere to put the step-down - unless on a separate board.

2) Power switch with high (relatively high!) resistance. Using a switch with a resistance of 0.05 Ohm would eliminate all problems with its heating and add about a watt of power to the cartridge heater. But the case would no longer be a 2mm dpak, but at least one size larger. Or even change the control to the n-channel.

3) Transfer ntc to pen. But then it makes sense to move the microcontroller, the power switch and the reference voltage there.

4) Expansion of firmware functionality (several sets of PID parameters for different tips, etc.). Theoretically it’s possible, but personally it’s easier (and cheaper!) for me to re-create it on some younger stm32 than to trample it into existing memory.

As a result, we have a wonderful situation - a lot of things can be remade, but almost any rework requires throwing out the old board and making a new one. Or don’t touch it, which is what I’m leaning towards for now.

Conclusion

Does it make sense to switch to T12? Don't know. For now I'm only working with the T12-K tip. For me, it is one of the most universal - both the polygon heats well, and the lead comb can be soldered/unsoldered with an ersatz wave, and a separate lead can be heated with a sharp end.
On the other hand, the existing controller and the lack of means for automatically identifying a specific type of tip complicates working with the T12. Well, what stopped Hakko from putting some identifying resistor/diode/chip inside the cartridge? It would be ideal if the controller had several slots for individual settings of tips (at least 4 pieces) and when changing tips it would automatically load the necessary ones. And in the existing system, you can, at most, make a manual selection of the tip. Estimating the amount of work, you realize that the game is not worth the candle. And the cost of the cartridges is comparable to an entire soldering station (if you don’t buy the ones from China for $5). Yes, of course, you can experimentally display a table of temperature corrections and stick a sign on the lid. But you can’t do this with PID coefficients (on which stability directly depends). They must differ from sting to sting.

If we discard the dream thoughts, the following comes out:
1) If you don’t have a soldering station, but want to, it’s better to forget about 900 and take T12.
2) If you need it cheaply and you don’t really need precise soldering modes, it’s better to take a simple soldering iron with power adjustment.
3) If you already have a soldering station on the 900x, then a T12-K is enough - the versatility and portability are excellent.

Personally, I’m happy with the purchase, but I don’t yet plan to replace all the existing 900 tips with T12 ones.

This is my first review, so I apologize in advance for any roughness.

Good day to all. The eternal struggle with the toad forces people to do unpredictable things. This happened this time too, and instead of a ready-made soldering station, I purchased a do-it-yourself kit. See what came out of this below.
From reviews on Muska I learned about the existence of Hakko T12 sting cartridges. This question interested me and, having started to study the information, I came across the set under review. After reading reviews and watching several videos, I realized that as a result you can get a pretty good soldering station for little money. I’ll make a small digression right away - to get a working soldering station for this kit, you need to ADDITIONALLY PURCHASE a 12-24V power supply. Naturally, 24V is the most preferable option, in which the potential of T12 cartridges will be fully revealed.

Table from the seller's website

So let's start - I was lucky and the parcel arrived in just 12 days. A gray bag wrapped in tape containing a cardboard box, inside there were small parts in separate bags. Everything arrived intact.
Parcel contents:

• The soldering iron handle is glossy plastic, the quality is mediocre. I asked the seller to put it in blue, by default a black pen is included in the package;
• Wire 100cm long, diameter 5mm, silicone, heat-resistant, does not remember shape;
• The first bag contains a soldering station controller, a red LED, an SW200D vibration sensor and an encoder knob;
• In the second there is an aviation connector;
• The third contains a kit for assembling the insides of a soldering iron handle;
• Bundle of wiring, harnesses and cambrics;
• The T12-VS2 tip also contacted the seller in advance and asked to replace it, because... By default, the kit includes a tip of type T12-K;
• Gift tweezers of passable quality;
• A note from the seller with a promise of goodies for subsequent orders))).

Well, the contents of the parcel have been reviewed from all sides, “sniffed”), and we proceed to assembly. I started assembly with the insides of the handle. And if you carefully read the title of the review, you already realized that you can’t do without a soldering iron. There are several nuances when assembling the handle, which I will now talk about.
1. There is a difference in how you orient the halves of the internals to each other; you need to do this in such a way that the areas for the soldered contact “curl plates” are opposite.


2. Through trial and error, I found out that the contact plates need to be soldered with the curls inward, this is not obvious from their shape, but believe me - it will be better and, probably, more correct. Because in this case, they are simply soldered in the middle of the contact pads and then contact the tip in the right places without any problems.




When soldering the lower part, you need to immediately decide on the wires and solder the wires while simultaneously fastening the insides of the handle.

Connection diagrams:

Solder capacitance 104 (0.1 µF) and vibration sensor SW200D




Solder the wires from the aviation connector side


Assembling the pen


This is what happened after assembly:


Now let's move on to the controller. Dimensions 67x24mm. The depth together with the encoder is 25mm, in the housing it protrudes 13mm.
And ours is quite smart and, in addition to its immediate responsibilities of regulating and stabilizing the temperature of the tip, it can fall asleep and turn off after a certain amount of time (which can be changed).

Photo of the controller

In addition, you can change the temperature control step settings and perform software temperature calibration. These parameters can be changed directly while the soldering iron is operating - modes P10 and P11. This is done as follows - press the encoder knob and hold for about 2 seconds, get to point P10, briefly press to change the order (hundreds, tens, units), turn the knob to change the value, then press again for 2 s. we hold the encoder knob, the value is saved, and we get to point P11, etc., the next 2s. pressing returns to operating mode.
But that’s not all, if you apply power to the controller while holding the encoder knob, you can get to a more advanced program menu. In the discussion of one of the video reviews, I found the following information on it:
P01 ADC reference voltage 2490 mV (TL431 reference)
P02 NTC setting 32 sec
P03 op-amp input offset voltage correction (55)
P04 Thermocouple amplifier gain (270)
P05 proportional gain PID pGain -64
P06 PID integration gain iGain-2
P07 PID differential gain dGain-16
P08 auto power off after 3-50 minutes
P09(P99) reset settings reset
P10 temperature setting step
P11 thermocouple gain (temperature calibration)
Temperature calibration took me quite a lot of time, but as a result I was able to achieve quite acceptable results.

Tip temperature measurements

Further assembly of the station very much depends on what power supply you decide to use, there is also one caveat here: when using a power supply of 19 V or higher, you need to unsolder the 101 (100 Ohm) resistor.


An LED and a male aviation connector are also soldered into the controller.
I used a fairly large 24V, 4A power supply. Therefore, the controller was installed directly into it. The result is a fairly convenient and compact device.

Power supply characteristics

Ready soldering station:


T12 stings are my kit. The last photo is very interesting; it clearly shows the difference in the logos of tips ordered in one store at the same time. I assume that both stings are fakes. But this does not affect the work in any way. Perhaps time will tell. If there are experts, I would be interested to hear your opinion:

It is difficult to draw conclusions regarding this product because everyone will get a soldering station with its own power characteristics (depending on the power supply) and appearance (depending on imagination, diligence, etc.) Therefore, I will only talk about what happened I have.
Pros:
1. Quick heating to operating temperature in about 15 s. Personally, I like the heating speed the most. Turned it on and while you take the soldering iron with one hand and the solder with the other, you can already solder.
2. Good power - you can warm up large landfills.
3. Temperature reset to 200 degrees (falling asleep) and self-shutdown after a certain period of time.
4. Heat-resistant wire, which can also be considered a disadvantage due to its massiveness and some elasticity. But for me, the thermal stability outweighs the inconveniences described above.
5. If you get the hang of it, you can change the tips without waiting for them to cool down, I got the hang of it - so that’s a plus)))
6. Well, naturally, among the advantages we will include the pleasure that a person gets from doing something with his own hands, especially when it turns out well and is pleasing to the eye.
Minuses:
If you're being picky, there are also a lot of downsides, including the mediocre quality of the handle and a fairly long tip reach. But for myself, I definitely singled out only one.
1. Out of the box, the temperature of the tip does not correspond to reality; I had to tinker a little to get an acceptable result. But even after calibration, the temperature fluctuates: at high temperatures it is lower than what the controller shows, at low temperatures it is, on the contrary, higher.

Conclusion:
If you have an unnecessary power supply and don’t have a good soldering iron with temperature stabilization, definitely take it. But even if we consider the issue of additionally purchasing a power supply, it turns out to be quite a good option.

This is my first review, I wrote it mainly at night in low light conditions, so the photos didn’t turn out very well. If you have questions, write, I will help in any way I can.

For my birthday I was given a soldering station with replaceable tips HAKKO T12. The kit included three tips, of which I use 2, and only because of poverty. Now we managed to take a set of stings for review - 10 pieces.

What are the benefits of this type of sting? Firstly, they heat up quickly - they heat up to operating temperature in 12-15 seconds.
Secondly, there is a built-in temperature sensor. If you have a normal soldering iron controller and an external temperature meter, it is possible to adjust it within +-7-10 degrees.
Thirdly, they are quick-release. Replacing one tip with another takes 5 seconds.
Fourthly - assortment

Of course, the Chinese brothers make copies, generally of good quality.

Why do you need such a set? Due to the wide range of parts, it is necessary to keep a wide range of tips. There is a universal type - but of different sizes, there is one for soldering massive parts, a needle type - for small SMD parts, a poker - where it is inconvenient to get to the part...

As a result, if you solder different types of parts, you end up with 5-7 tips, which you use often.
But let's get back to the set.

It arrived in this form, packed in a cardboard box and bubble wrap.

The tips have 3 contacts separated by plastic rings.
The length of the tip in the set ranges from 147 to 154 mm - depending on the type.
Each tip has a sticker with the tip type and code.
Tip diameter 5.5 mm
Supply voltage - 24 volts
Power 70 watts
Temperature - up to 400 degrees (up to 450 is possible - but service life is reduced)
Compatible with lead-free solders

The set contains the following tips:
T12-B
T12-BC2
T12-D4
T12-C1
T12-C4
T12-D08
T12-D24
T12-IL
T12-JL02
T12-K

T12-K - convenient for heating several contacts or a massive part, for non-standard ones - welding polyethylene or cutting synthetic fabric.

T12-D08, similar in shape T12-B and T12-IL differ in diameter and sharpening angle

T12-JL02 - used in hard-to-reach places

T12-D4, T12-D24 - Chisel sharpening

T12-BC2,T12-C1,T12-C4 “hoof” - diameter 1, 2 and 4 mm universal tip sharpening

All tips came with a tinned tip.
They solder well, when soldering with ordinary rosin at a temperature above 300, black carbon deposits form on the tip, it is better to use specialized fluxes.
Personally, the kit lacks a “microwave” tip and one with a recess for soldering lead elements.
After a month of use, I did not find any traces of burnout on the sting. The copper one would have to be sharpened twice already.

Nice set for a reasonable price.

The product was provided for writing a review by the store. The review was published in accordance with clause 18 of the Site Rules.

I'm planning to buy +24 Add to favorites I liked the review +13 +31