Skip to main content
Topic: Really universal soldering controller (Read 773923 times) previous topic - next topic

Re: Really universal soldering controller

Reply #2625
lm2675 has an extremely simple application circuit, if it does not work most likely it is a fake chip, do not buy it from aliexpress, only buy from mouser, digikey or farnell, it is expensive, but otherwise you will lose money, time and frustation.

Re: Really universal soldering controller

Reply #2626
yeah I had them from aliexpress because I couldn't find any source that is selling here and does sell to private people and had them.
Mouser and Farnell don't seem to supply to private people and the farnell site for private buyers develektro didn't have them.
Digikey would have had 18€ shipping so ordering a single chip there is pretty expensive :D, did not know that one either when I ordered the inital part list :), next time :D
This is like my first time building something like this, so I don't really know too many places to order them from :D.

But yeah yesterday I found them from a more trusted source reichelt (did not know this one before) after figuring out that I apparently could use LM2674 and LM2675M-ADJ instead of the LM2675M-3.3 so I ordered some of these. Guess I am gonna wait for them to arrive :)
Thank you for taking a look.

Re: Really universal soldering controller

Reply #2627
mouser and farnell both sell to private persons, like digikey. There is also TME.

Re: Really universal soldering controller

Reply #2628
Well I put new LM2675M-ADJ and 1.8k 0.1% and 3k 0.1% Ohm resistors now and replaced the NRVTSA4100 with SK310 SMA (replacing the diodes did not change anything in my measurements).
And now when nothing is connected to Back-PCB I get 3.3V and -0.5V.

But if I connect the Front-PCB it changes to ~1.2V and -0.5V.
No idea how I should go about trouble-shooting this :(

Re: Really universal soldering controller

Reply #2629
But if I connect the Front-PCB it changes to ~1.2V and -0.5V.
No idea how I should go about trouble-shooting this :(
I see only two options: something is still wrong with the regulator (or it's input) and it's not providing required current, or you have a faulty component on the front board that pulls all the current/some short (not so short actually) circuit.
For option one: measure voltage at C7 under load. There should be a voltage slightly lower than Vin. Do not measure on U1 directly as leads can easily slip and short input to output. Connect a load to the regulator to pull around 200mA of nominal current (10-15Ω 2W+ resistor), and check the output voltage.
For option two: as you don't have near zero voltage, around 1.5W (LM2675 has 1.25A current limit) of power is dissipated somewhere and can be easily detected by a finger test.

Re: Really universal soldering controller

Reply #2630
For option one: measure voltage at C7 under load. There should be a voltage slightly lower than Vin. Do not measure on U1 directly as leads can easily slip and short input to output. Connect a load to the regulator to pull around 200mA of nominal current (10-15Ω 2W+ resistor), and check the output voltage.
:) I got only a bit above 3.7V at C7 under load, at R2 it has ~35V on one side and the ~3.7V on the other under load. (no load it is 35V and 15V)
Gonna try and investigate why that is happening there. Thanks so much already, man I was already pulling chips of the front board one by one to see which circuit might cause it :/ now I can at least stop that :D

Re: Really universal soldering controller

Reply #2631
:) I got only a bit above 3.7V at C7 under load, at R2 it has ~35V on one side and the ~3.7V on the other under load. (no load it is 35V and 15V)
Maybe D5 is reversed or faulty? Otherwise the problem should be in R2 or ZD3.
Minkok is pointed out the input problem on the previous page, you should have tested it in the first place.

Re: Really universal soldering controller

Reply #2632
Maybe D5 is reversed or faulty? Otherwise the problem should be in R2 or ZD3.
Minkok is pointed out the input problem on the previous page, you should have tested it in the first place.
Yes I did not understand his sentence ,
Voltage at a pin 7 should be close to Vin (some forward voltages less )
Was quite confused about what Vin was referring too, can only shake my head about myself at this point becaue I thought he was just referring to the PIN named VIN (Pin 7) of LM2675. No idea how I did not realize he meant Vin of the schematic. :( So I was just majorly confused why I should compare voltages at the same point :(

And then after I figured that the LM2675 I was using where faulty, I didn't read his post again, hugh error on my side because he actually referred to D5 too. Which I changed now and everything seems good from the voltage outputs, even with the Front-PCB connected...
12V (~14V) is there 3.3V is there -0.65V is there.


Re: Really universal soldering controller

Reply #2634
Even maybe LM2675 was not faulty after all...
Yeah I actually gonna test them again once I got everything running, just to find out :)

But for now I after I resoldered all the chips that I had removed back onto the Front-PCB I had a short to ground that I got fixed after a while and then I had 13 on the display so I redid all the chips hat used I2C (since some one mentioned that in an earlier post) on the front board and noticed finally the very little marking on  U11 so now that one is soldered on the right way round.

Now I am stuck with interchanging 21 and Power Lost which gets displayed if something is < 0.5A and < 7.4V . Gonna need to figure out what through after I slept :) but things are going somewhere finally :)

Re: Really universal soldering controller

Reply #2635
So after I figured I actually connected CNTR the wrong way round, I now get flashing  21, I can see 300° and then POWER LOST.
So I guess I sent down 24V the SENSB line, any easy way to figure out if I fried something there along the way :/
The circuit around U10 should be fine since D18 should have blocked it, If I can read that right.
But no idea if it could have done something to the circuit around ADG734 and not sure how to actually find that out.

Other notes:
Across ZD2 and ZD1 I measure ~4.2V when the board is powered.
Across ZD4 I measure ~3V when board is powered. (From what I found by searching in this topic about my problem, when measuring this with DMM the value is lower then peak voltage)

I used SZMMSZ4683T1G which 3V Zener-Diode I got from develektro, the website farnell sends you to when you try to buy as a private person in germany. All the items apparently come from farnell in the end. Markings are CJ like other people but they are the right markings according to the Datasheet :(.
But getting these values lets me assume those Zener-diodes all don't work and I should get replacements?

Re: Really universal soldering controller

Reply #2636
But no idea if it could have done something to the circuit around ADG734 and not sure how to actually find that out.
It should be OK. Sense inputs are rather high impedance and clamped (if you have working and correctly placed diodes of course, ADG734 should have its own steering diodes though).
You've send HSEL to the ground. MCUs don't like short circuits on its outputs. As the second channel is enabled for a brief period of time without dual channel tool connected it may be OK.
Just supply controller from any DC voltage source of 12V or greater, and carefully check ZD4 and R51 voltages. The voltage on ZD4 may be slightly lower as current through it is small. On R51 should be Vin/28 volts.
And check ADC reference voltage by the way.

Re: Really universal soldering controller

Reply #2637
It should be OK. Sense inputs are rather high impedance and clamped (if you have working and correctly placed diodes of course, ADG734 should have its own steering diodes though).
You've send HSEL to the ground. MCUs don't like short circuits on its outputs. As the second channel is enabled for a brief period of time without dual channel tool connected it may be OK.
Just supply controller from any DC voltage source of 12V or greater, and carefully check ZD4 and R51 voltages. The voltage on ZD4 may be slightly lower as current through it is small. On R51 should be Vin/28 volts.
And check ADC reference voltage by the way.
Powered it from a 12V 6A DC power supply and it started up and showed "No Instrument". Voltage at R51 was indeed ~Vin/28 (0.425).
Voltage at ZD4 is ~1.975V. and across ZD2 and ZD1 it's ~3.6V.
Vref measured at C42 is 3.00*V

This just makes me more confused why it isn't working with the toroidal 160VA Transformer as powersource :(

Re: Really universal soldering controller

Reply #2638
With the 50hz AC input from the Transformer Vin is 23.17
Voltage at R51 is 0.875 (pretty close to Vin/28)
Voltage at C42 is 3.009 (it measured 3.008 with the AC supply, but really I think the last digit isn't even anything my MM can measure correctly in the first place)
Voltage at ZD4 is is 2.947V
Across ZD2 and ZD1 is 4.26V

I have no instrument connected, so when it says that with the DC power that is alright.

Re: Really universal soldering controller

Reply #2639
Voltage at R51 was indeed ~Vin/28 (0.425).
Voltage at ZD4 is ~1.975V.
This looks good. On 24V AC it's also OK.
But ZD1 + ZD2 seems too low. It should be ~5.6V and across C12 should be around 5V.
I don't know exactly how Vin is measured and triggered while on AC. Looking at the code a pretty much complex stuff is happening.