SPOT WELDER WITH PIC16F88 tested - YOUTUBE video

[Ichan]

I have no answer, i didn't calculate anything too.

Just my feeling that it should not turn off too fast, as the very high welding current effected by stray inductance could kicking back to the mosfet destroying the body drain diode if the turn off is too fast.

How slow? Can it be too slow? Frankly, i do not know.

-ichan

[kalpakchiev]

I agree but we have to talk with dimensions - too slow or too fast sounds like the saying for comparing BIG ANT and SMALL ELEPHANT .

I have no idea which way is better to turn it off fast or slow.

[borberk]

100m capacitor charged on 30V and discharged through 0.07+rds/5 is practically empty in 30ms.
Troubles with voltage spikes vhich occur on weld start and end because of electrode inductance can be eliminated by diode in series with charge FET.
Rise and fall times on weld FET gates are not important.

[kalpakchiev]

100m capacitor charged on 30V and discharged through 0.07+rds/5 is practically empty in 30ms.
Troubles with voltage spikes vhich occur on weld start and end because of electrode inductance can be eliminated by diode in series with charge FET.
Rise and fall times on weld FET gates are not important.

Yes I think that's right. In our case capacitor is 1F and voltage is 20V so we are talking for max. weld time about 200ms

[titi]

Hi,

In the web site : http://www.ultrakeet.com/index.php?id=article&name=cdWelder_p2

The you-tube video show examples of weld, at 1:58 min they give time of weld of 0.1ms, 0.5ms, 1.5ms, 4.5ms, 9.5ms to weld (Nickel to Stainless).

If these times are really good, if you have not a very fast turn-off, it will be impossible to have these short times.

But it is possible that for your applications, that turn-off time become not important.

Best Regards.

[waicon]

a) yes, R9 increase will couse turnoff timing increase.
b) Q6 will it replace with P-ch FET (small signal to92) type? as long as its -VGS is higer than VIN. or added 12v ziner.
c) ring effect on gate ? i'm not sure, it related to capacitive load effect. refer to link http://rapidshare.com/files/378642574/capacitive_loading.rar , but i think R1~R5 slightly increase turnoff timing.
d) a good weld spot, are determine by a good "intermatalic formation" in its spot by 2 metals, if less will couse loose or "dry-join", too much will properly couse a melted hole in it. a PWM pulse train might help on weld input. 


 

[sadman]

Hi Friend kalpakchiev

can i used uln2003 in place of tc4422 also what is total current of transformer you used for supply and charging purpose

regards
sadman

[kalpakchiev]

Hi Friend kalpakchiev

can i used uln2003 in place of tc4422 also what is total current of transformer you used for supply and charging purpose

regards
sadman

I am not sure You can replace tc4422 whit ULN2003, transformer is 16VAC 240W with 35A bridge rectifier
Best regards

[pickit2]

waicon said a PWM pulse train might help on weld input.
I think back to a DC welder I fixed a few years back its control was with two power Fets.
one had PWM on its gate and other was on/off   

[kalpakchiev]

waicon said a PWM pulse train might help on weld input.
I think back to a DC welder I fixed a few years back its control was with two power Fets.
one had PWM on its gate and other was on/off  

Hi Pickit2, why there is an ON/OFF fet when turning PWM off, also turns the weld off?


How many ONs and OFFs in one post  
Best regards

[kalpakchiev]

LCD prepared for test 

[Ichan]

I am still waiting... 

Can not find my old tab welder, searching for the caps i found a bag of 10000uF/35V, is that a crazy idea to parallel 100 of them making a capacitor farm?

-ichan

[sadman]

I am still waiting... 

Can not find my old tab welder, searching for the caps i found a bag of 10000uF/35V, is that a crazy idea to parallel 100 of them making a capacitor farm?

-ichan

i found 47000uf used 24 in parallel to get the required value

[kalpakchiev]

It is possible to parallel a few capacitors to reach desired value (make a cap. farm), but you must have in mind that this will increase internal resistance of the whole capacitor.

Best regards

[Ichan]

Isn't paralleling capacitors will decrease the total ESR value?

-ichan

[kalpakchiev]

I think that all the connection between capacitors have to be added to internal resistance, so it will be increased. Am I wrong?

[sadman]

I think that all the connection between capacitors have to be added to internal resistance, so it will be increased. Am I wrong?

if you have ESR meter just do a test for us that will help us as in my country 1F audio capacitor hard find, just check ESR of two 2200uf,cap then join both capacitor in parallel and measure ESR again that will clear every thing so we used alternative method, once again thanks for your corporation.

regards
sadman



 

[kalpakchiev]

Yes this test can be very useful, but unfortunately I have no ESR meter.

If some one in the forum have, please help Sadman.

Best regards and many thanks in advance for your kind cooperation.
Ivan Kalpakchiev

[Fab66]

Hello kalpakchiev, all's

i am new here, i discovered this site and this project.
I follow it since a few day's, and i think that can be my
next electronic build.

Thank's for sharing, i search now the condensator on internet.
Do you think that best result will be with 2 condo in serie,
or one (with 1F) is enough to sold accus ?

Best regard's to all

Fabrice

[kalpakchiev]

Hi Fab66,
Sorry for my late post but I was on a place where it is difficult to get connected to internet.
For me using one 1F capacitor gives good results. But it depends on what You are going to weld - type and thickness of the material.
Now I am back and will finish the project with all the final modifications that we discussed with other peoples in this forum.

I wish you success in your project.

Best regards,
Ivan Kalpakchiev.