UHU Chip problems

[LZ1TWB]

Cheap IC sockets get easily damaged if you heat them more than enough. The spring tension in each pin can get loose or even the plastics to melt. I've had a problem with sockets and tend not to use the ones with the flat pins. Those with the round pins are always more robust to deal with.

In fact using sockets on a thing like this is not a good idea. Having transformers that vibrate and magnetic contactors that shoot like a gun in the same cabinet is just asking for trouble. On my 4 boards I use sockets but I have "glued" the IC's so they don't move.

Just an idea on bad connections.

Todor

[mibo1112]

Hi Todor,

And thanks for those ideas.
I always used sockets with round pins, but to be honest, if the temperature is high, and I usually use about 250 Celsius (which is pretty hi) the pins tend to get hot and melt plastic. So, I don't warm them too long - and that was probably why that bad connections occurred.
By the way, those pins are just inputted in the holes of a socket. If you ever need, each pin you can take out of the socket, by just pushing it on the direction where the chip should be. It is a little tricky, and can damage a pin or the socket. When you putting it back in the socket, it will make "click" sound. This way the socket is undamaged, well by the heat at least. Earlier I used to take the benefit of heating these pins for taking them out. But, as you mentioned, the socket will melt down.
To tell you the truth, never did thinking to glue the chips.
Probably the best solution, from connection point of view, is to solder chips directly on a board. Well, in that case it will be relatively hard to take them off if something unexpected occurs, and you have to change them. Vacuum solder will be needed... It can be done without it of course, but we are familiar with that job too, so...
Anyway, thanks again for the ideas.

Milos.

[cnc2]

Hi everybody !

Hi Todor,

And thanks for those ideas.
I always used sockets with round pins, but to be honest, if the temperature is high, and I usually use about 250 Celsius (which is pretty hi) the pins tend to get hot and melt plastic. So, I don't warm them too long - and that was probably why that bad connections occurred.

Milos.

And it's here that my special, super, amazing, custom anti ESD work station comes into play... first it protects the parts from ESD & Second it prevents you from overheating anything... I've successfully assembled my first UHU board using a cheap 40W soldering iron (cheap<=>3euros) plugged into a light dimmer with no temp control as you can see on the picture it just acts like a BIG heatsink.
Here you go !
The result is no melted out sockets nor overheated parts !

Thanks !
cnc2.

Don't forget to put a resistor in series with the wrist strap

[LZ1TWB]

I love the wrist strap, looks like my casio. How did you manage to hook the wire to the small screw? I guess the series resistor should be in the MOhms range in order not to shake you if you touch somewhere with high voltage.

Nice setup, the old-looking board brings me in the medieval ages. The next moment I realize we are assembling some real world servo controller though.

Todor

[cnc2]

Hi Todor !

I love the wrist strap, looks like my casio.

You've got it right, it's a Casio DATA BANK 150 & had great features...it's a dead watch, i've got it as a present but it didn't last so long because it had a dead thread so it wasn't water resistant anymore & i've discovered this in the first summer when i've got into the the water (at the beach) before even i start to swim ...yet another "water NON resistant" watch..haha.

How did you manage to hook the wire to the small screw?

Well, it's a dead watch so, i've removed a screw & hooked a phone wire (0.6mm² not the one in the pic, because it was to big & weak) to the cover plate through the hole left by the screw...the plate is made of stainless steel so, you can't solder on it...you'll have to pass the wire between the plate & the plastic so it's squeezed tight & won't break the hooking of the hole (I hope you understand what i mean).

I guess the series resistor should be in the MOhms range in order not to shake you if you touch somewhere with high voltage.

You're right about the resistor it should be around 1 MegOhm from this document posted by Kreutz, even Henrik warned me but I had already tasted ~40V of AC juice with no resistor hopefully the path was short enough (finger to wrist) so my arm didn't shake & everything went ok.

Nice setup, the old-looking board brings me in the medieval ages. The next moment I realize we are assembling some real world servo controller though.

Todor

Thanks, I've used your trick to increase the pcb current rating, I've solered some solder wick on the high current side of the board & it works great, you can't beat the quality or performance to price ratio of those old boards they should be enough for Hobby use, the HP UHU is for the bigger machines (industrial Hobby) which justify the need for power & price, a really good concept...the industrial revolution at the hobby level

Did you finish building your machine ?

Thanks !
cnc2.

[LZ1TWB]

Yes, the machine is "almost" done as we like to say. The problem why it took me so much time is that firstly I tend to do everything the hard and complex way, while people do it simple and make money in the mean time. The other thing is I am doing so many other things.

Anyway I am working on the spindle regulator, which is now getting mounted in a box. I have started a thread at PWM drive for a 2KW universal motor/router but there is not much interest of the design so I struggled my head alone.

To say something on the topic I found the solution of reseting the UHU when there is a wire connected to the E-STOP IN pin. It seemed that some noise is getting in there and when you twist some of the motors and the current exceeds some (very low) value the chip is reset. Very annoying in the first place but was gone when I put a small cap between the pin an ground. At the beginning I tried with a very small one to see when it will take effect but it seemed that even 27pF solved the issue. So I put 1nF to be sure. I guess this is a problem of the schematic, as there is a 22Kohms pull-up but it seems inefficient about this.

Todor

[cnc2]

Good news, could you post some pics of your machine ?

For the spindle regulator, I have to say that I don't have this kind of knowledge, I'm a computer/network guy you know....Btw why are you struggling to build your own regulator while you can buy a cheap Chinese spindle+VFD for around $500, some people on the zone are using these & are happy with'em...Maybe you have a really good reason ? let me know, please.

Thanks for giving me the solution for a problem i'm not facing yet, I thought that a shielded wire would be enough to prevent any problem & could never think about using a capacitor to fix it, i'll remember this one.

Thanks !
cnc2.