Using 5M instead of 10M doesn't give what we want,only increased noise and a delay when touching changing tip part(at least for me and once again I am using the sensing board inserted in the unisolder box not in the stand-don't know if it matters) so what I did was to keep the 10M resistors; I have changed the 2.2 micro capacitors with 1 micro to decrease delay at inserting and taking out the handle too but the doomedahab effect at boot was still there.Because I was satisfied with the changing tip part now,I have rotate the resistors only on the sleep part and used 165k for R2* ,20k for R5 and I am happy with the results.Act as it should at boot,got around 0.4-0.5s delay at wake from sleep,responsive at tip change,stable,doesn't react at hand touch.I think who is not satisfied with the delay can try an 500-680nF capacitor for C4. So,what worked for me was: C4,C5-1µF;R1,R4-10MΩ;R2*-165KΩ,R5-20KΩ;R7-330KΩ,R5*-20KΩ
It is the wrong way to change C4 and keep the same R ( it is the filtering part ) . change the R ( you will keep the same Xc and decrease the time constant and be aware of the increasing the MOSFET influence )
i get a spike every few seconds on j4 and j8. Forgot to check on heater 1. Some werd other stuff sometimes if i probe with multimeter or oscilloscope on some points the Iron starts heating at low power sometimes 1% sometimes bit more.
Can you please check HEATER1 and repeat the same way with Vout1+ and Vout1- ( now you know for what to look), to be sure it is the fet's driver
@Ne_0, you can try also the same R1&R4 - 5M Ohm and R2&R5 -330k
@Ne_0 sorry, I made a mistake - You need to change R1&R4 - to 5M Ohm and R5&R7 leave as is 20k, and R2*&R5* as is 330k, to be able to get the difference b/n the counted and the real delay ( I really sorry I did not change on the Sch old marked R* resistors). You do not need to take a video, and just notice if it is faster ( on pick up ) than the Sch with 10M Ohm resistors ( you mentioned a 1.5 to 2s delay ) and check if it works fluently or if the noise is increased. I want to be understood correctly, It needs to be done with good filtering, but not at the price of a delay like that.
Because: the "tip change" message appeared with the NP105 nano tweezers connected (and in the holder), after putting the not connected T245-A handle in the holder and taking the nano tweezers out of the holder.
So i think i'll stay with the limited filtration for now.
Can you try to place in parallel another one 10M Ohm to R1 and R4 ( to get 5M Ohm altogether) for both
When I change the 330k to 20k and vice-versa,its working as it should at power on,but there is a 2s delay when I take the handle from stand until its boosting the temp.I tested with a 245 handle and I use the sensing board on the back board of the unisolder not in the stand.
With 330k ( no noise ), it should be around 0.5s - discharging C4 ( on pick up), and 0.6s charging C4 ( back to stand). But if you want you can change it to 220k or to already tested by Doomedahab 150k.
I have the same effect with 330k resistors. It only happens, when the iron is in the holder from the get go. If the iron is outside of the holder while switching on my Unisolder and i put the iron in the holder while it heats up, it "clamps" almost immediately (maybe with a small delay around half a second).
Still have to test the new configuration...
You use "Test Config". It is definitely from not good filtering. What kind of tool do you use?
I did not notice any interference with your design, but there is a significant other effect when I switch around the resistors: In the "old" configuration, when I would turn on the controller, the temperature would go up to almost the set temp before the holder would "clamp" the iron and reduce the temperature to the holder temp.
In the new configuration, the holder "clamps" the iron immediately.
By the way, I use 20K and 150K resistors instead of 20K and 330K so it might be part of the reason why in the old layout, this effect was observed.
I think the mentioned effect might be from the noise generated by unisolder itself (fast heating and lack of enough output filtering in the test config sensing board), by the way, what tool did you use, T210? The new config is better, isn't it? Using 150k instead of 330k makes the filtering weaker (7 times better from test config), but with a faster reaction to states changing around 0.3s , corresponding to 0.7s for 330k
Hi again. I redesigned the contact sensing board. Hot, direct from the working table, Now it is much simpler and I expect to work more smoot. Again it has 2000V protection ( static and peak ). Use 20V GS for the BSS138 version. That sch has not so strong output filtering compared to the last and is more stable. Unfortunately, it is not tested, use at own responsibility, the time someone checked it and confirm. Now I publish only one board and sch
Hi, Good news, after around one year of testing by all of you in different countries under different conditions the "Test Sch" is prooved to work - the results prove, that after the protection circuit passes only sporadic noise and even limited filtration does the job. Now is time for full filtration ( you Do Not need to change the Board), just to rotate two resistors, as is in the picture. Of course, if you want. Thank you for your trust.
So now for the measurements. With the Iron and heater connected. I get around 0.2V AC measured on bot Vout1+ and Vout1-. Also for if It matter when i go to info menu i get Cjtemp - 273 and a roomtemp of 14(roomtemp seems okeyish)
Seems Q2 dos not work - check, please ONOFF at J4/6p and J8/6p connectors with an oscilloscope and HEATER1 to see if PWM passes ( it should be for less than a second I suppose - pay attention for that)
so essentially It is a tindie board. So of course there is a Problem (i would not be here otherwise) when i connect a c245 It gets recognized properly and says heater open. But when i connect the cartridge in the Iron It still says heater open. The werd thing is i got It work for a moment did not use It for a week and now It is having the Problem again.
Asuming all the rest is ok, it should be easy. It will be good to have an oscilloscope if we do not get straight results, You have a problem with power MOSFETs or their driving (if heater connections are ok). Start point - Measure with the multimeter (AC) the voltage b/n GND and ( Vout1- and Vout1+) - this will give you an answer if MOSFETs are working (but not if they are ok) and the direction where to look for. Work carefuly!
ps; Q10 and its driving should be ok ( you get the right ID), but do the measurement, in case of firmware corruption)
You are welcome, and sorry I did not catch the joke. I also started with one tool and 2x cartridges ( original JBC ) and now I have many of them ( mixed ). You will spend much more money with pleasure when you built it! That project is based on precise measurement, you need to use proper parts.
For 100-150 USD ( if they are important ), you can buy from China good iron, supporting the JBC handle, but you need to count how many tools are supported ( depending on your needs ) and what will happen, in case of its failure.
Hi Guys, I joined the forum for this great looking project, linked from Marco Reps. I was considering buying a JBC, when my DIY demon started whispering in my ear that that was a lot of money for something I could build myself for less than 10x the price of a fully assembled model, so why not? .
You should expect sooner half of the price ( depending on your bin ).
My question is (and I apologise for not reading the whole thread first) Is the bom covid current? or am I better off designing my own PCB? That way I could also make sure to use as much of my part bin before buying more parts, but If the Bom is fairly current i'll have a bash at loading it into Mouser and see what this will cost me. probably not a great deal i'm guessing, given the quality, given I have a fair whack of the BOM, looks like it'll be under $100 plus the iron. While I will certainly go that way if required, Id like to avoid designing the PCB is possible, I have too many projects on right now as it is, while soldering it up looks pretty straight forward.
Yep, by the time, but not by the price - they change with a delay. !00 USD with iron is a sweet dream ( even with a Chinese one ), You need an extra Transformer, case, cables, connectors, stand, and so on. Count again
. change the R ( you will keep the same Xc and decrease the time constant and be aware of the increasing the MOSFET influence )
.