What NOT to do with the HP-UHU...(self inflicted)

[H.O]

Hello,
If you've been following Irfan's UHU saga you might know that I had a little dissaster happening the other night. If you haven't been following this is what happend.....

In an effort to help Irfan solve the issue with invalid transitions on the encoder input that we've been suffering from with the HP-UHU i rigged setup with the motor clamped to the bench by a big C-clamp (as I've done many times before). Basicly all that happens is that Mach3 cysles thru a G-code program and I que the UHU for status once in a while. After running nicely for 20minutes I decided to let it run and went inside to watch some TV. Picture 1 shows what it looked like when I left it.

After about 45 minutes I went out to check on it and found the motor (19kg servo-motor) on the floor (not running). It must have "escaped" fro the C-clamp and trown itself over the edge during a rapid direction reversal. The capacitorbank and dumpcircuit was also on the floor and the drive was hanging from the edge of the bench suspended in cables - it was still powered ON but not working....

I don't know how but the motor seems to have survived. It turns freely and the shaft runs true, even the encoder still works as does the logic part of the drive. The output-stage on the other hand is not in good shape.....

In the fall the motorcable shield had made contact with one of the motor output terminals resulting in one partly melted connector (picture2). When removing the heatsink I started to realise just how much power there had to have been in that short, see picture 3.

Picture 4 and 5 shows what the PCB looks like....

I'll try to clean the PCB with a fibre glass brush in an effort to remove all the charring, then I'll install a new MOSFET (the other seems to be OK) and a new drive chip and solder a think wire from the remaining trace to the MOSFET leg. Hopefully that will bring it back to life.

Anyway, nothing to blame on the HP-UHU just a little story about when things go wrong.

/Henrik.

[contactirfu]

Henrik, sorry to see the drive in this state, btw what are those heatsink pads you are using.

where can I buy them? are they available online anywhere?

[contactirfu]

To add to the topic - never switch the motor power supply first - always switch on the logic first and then the motor powersupply. If possible add a delay switch in between.

and when assembling a KIT make sure you have the 35V tantalums and not the 25V ones.

using 25v tantalums burnt a hole in my HP UHU PCB. (much worse than henrik's present situation) It was scary!

RGDS
IRfan

[H.O]

No worries Irfan, it was nobodys fault but my own, new MOSFETs are in so I I'll have a go at fixing it later this week. I was much more worried about the motor but it seems to have survived the fall remarkably well.

You can find the datasheet for the heatsink pads here Not much data there, I buy them from a Swedish supplier ELFA but they should be available "anywhere". Serach for silicone rubber insulators etc.

/Henrik.

[LZ1TWB]

Originally Posted by H.O Hello, The capacitorbank and dumpcircuit was also on the floor and the drive was hanging from the edge of the bench suspended in cables - it was still powered ON but not working.... /Henrik.

Henrik,

So you say it was still powered ON? I assume you don't have a fuse there or it was too high a value to trip. This is something not good and it's a miracle it didn't set on fire.

Anyway, from seeing how you've used your C-clamp on the first picture, surely that was not good enough. You should have clamped the motor mount parallel to the table with the motor hanging outside it.

I also have a similar motor in size and know what happens when I turn the acceleration in MACH in the range of thousands. That's why I tend to keep it on the floor, so it just can't fall no more.

So I guess this is a part of the experience one should take during the build of a diy-cnc machine. I've also burned a power stage by a stupidity once and now I know to be more carefull with those volts and amps hanging there.

Regards,

Todor

[H.O]

Hi Todor,
I agree, clearly it wasn't good enough - hard to argue with that.... But I had been running it like that all the time without problems but of course the minute I leave it out of sight it decides to escape.

I DO have a fuse but in this temporary powersupply I put it between the rectifier and capacitors to allow large currents while accellerating while still protecting the fairly small rectifier and transformers. I was fully aware of what would/could happen IF the DC-link was shorted but you know - that won't happen... yeah right... ;-) In this case the PCB trace going to the MOSFET drain acted as a fuse and vaporised. Judging from look of the heatsink it probably was on fire, for a very short time :-)

On the machine I have a fuse between the rectifier and capacitors as well as one for each drive with anti parallel diode across it.

The drive is now fixed, the only thing busted was that single MOSFET. A thorough cleanup of the PCB and a bit soldering fixed it. Tested it hard for alomost three hours straight and it diodn't miss a beat.

Every day is a learning experience!

Sincerely,
/Henrik.

[kreutz]

A short portion of the power copper traces on the PCB, near the Mosfet's leads, were designed as safety fuses, they are supposed to blow (a couple of millimeters) when the current exceeds 60 Amps for a few milliseconds...

[contactirfu]

Kreutz - SIR you are the greatest!

[cnc2]

Hi everybody,
Hi Henrik,

I'm going to assemble my UHU boards and found that for soldering MOSFETs, an "ESD safe" (electrostatic discharge proof) soldering iron should be used, as i'm a beginner i'm not sure if it's that critical to use such a soldering iron for assembling the UHU board ?

When you soldered the MOSFET on the board did you use an "ESD safe" (electrostatic discharge proof) soldering iron ?

Thanks !

cnc2.

[kreutz]

Originally Posted by cnc2 Hi everybody, Hi Henrik, I'm going to assemble my UHU boards and found that for soldering MOSFETs, an "ESD safe" (electrostatic discharge proof) soldering iron should be used, as i'm a beginner i'm not sure if it's that critical to use such a soldering iron for assembling the UHU board ? When you soldered the MOSFET on the board did you use an "ESD safe" (electrostatic discharge proof) soldering iron ? Thanks ! cnc2.

Most of the soldering irons sold, since the mid nineties, already have the tip grounded (they use three prong cable plug), they are inherently ESD safe if you don't cut the ground pin, and the receptacle wiring is correct. Normal ESD precautions should be taken during the whole board assembly if the relative humidity is 60% or less.


Kreutz.

[cnc2]

Originally Posted by kreutz Most of the soldering irons sold, since the mid nineties, already have the tip grounded (they use three prong cable plug), they are inherently ESD safe if you don't cut the ground pin, and the receptacle wiring is correct. Normal ESD precautions should be taken during the whole board assembly if the relative humidity is 60% or less. Kreutz.

Thanks for your reply Kreutz !

As i'm usualy using a cheap chinese soldering iron (with two prong cable plug) and i never saw a soldering iron with three prong cable plug in store...& i don't have a grounded wall receptacle in my room the only grounded wall receptacles are in the kitchen or in the bathroom... so please, if you have a way out of this issue it will really help me.

(remember the UHU board is the biggest & most expensive project i ever started)

Thanks !

cnc2.

[kreutz]

Originally Posted by cnc2 Thanks for your reply Kreutz ! As i'm usualy using a cheap chinese soldering iron (with two prong cable plug) and i never saw a soldering iron with three prong cable plug in store...& i don't have a grounded wall receptacle in my room the only grounded wall receptacles are in the kitchen or in the bathroom... so please, if you have a way out of this issue it will really help me. (remember the UHU board is the biggest & most expensive project i ever started) Thanks ! cnc2.

What is the standard in your country, 120V/60z or 220vac/50Hz?