I know that I have not been on this forum for a very long time life has gotten in the way

sam_sam_sam · 2018-04-04 02:53

This for the people on this forum when I spend lot time on here

I have have not done any Basic Stamp projects for a very long time

I now work for a metal supply company as all round maintenance person which means that I work on just about anything and every thing
from overhead crane to plasma cutting bed table CNC machine to a CNC drilling/ plasma machine to computer stack controller band saws
to a blever machine

I also do some outside consulting work repairing CNC servo controller power supply modules and other switching power supply and motor indexer controller

Sometimes I can become very busy at times

Right now I am working on a motor indexer controller which is giving me a big headache
About a year or so ago it would not remember where it was and could not find home and so on

It has a battery back up memory which the battery went bad and had very bad case of bad capacitor ides
change the memory chip and capacitors that had a very high ESR value reading compared to a new cap of same UF value

After this repair it work great for a little over a year then I get a call that it blew apart very small value cap (104) and blew the fuse for the power supply for the servo motor I did some checking and found out all the high power mosfets are being turned at the same time ( it has FW/REV directions) now if you take out the four (optic/mosfet ic chip) the short is gone but as soon as you put two of the (optic/mosfet ic chip) back you have the short again now the (optic/mosfet ic chip) is controlled by another chip that is controlled by the micro controller

The problem is that the only place I can find this chip is from China which will take about a month to get here unless I want pay a lot of money for faster shipping

One note this indexer is from the 1970’s or 1980’s and is no longer made and the company that makes it was bought out by another company which does not really support it anymore parts are very hard to find ( ic chips )

This is just one of the problem project that I have going on right now

That all folks for now

erco · 2018-04-04 04:05

Good to hear from you, Sam!

We all get busy at times. It's always nice to come back here to the forum and be among friends!

Genetix · 2018-04-04 04:07

Have you tried NTE for a replacement chip?

Peter Jakacki · 2018-04-04 04:20

Nice story Sam, but you gave us no hard information to go on to help you. Part numbers, drawings, photos etc?

frank freedman · 2018-04-04 05:47

Put four channels of a logic analyzer on the opto inputs and see if any drive phases overlap or don't change. If not check that the driver stage to the gate pins do not have a driver stuck on causing the stack to have both FETs on for a rail to rail short when the good drive says turn on. AMX-4 rotor/collimator lamp uses two IRF350S in totem pole arrangement. If one gets punched usually both get taken out. Worst ever was one board that smoked every trace, transistor and passives back to the opto outputs. Schematic can be had from GE common document website for xray devices. Has been in production since late 80s or so, running from an 8749 mcu.

sam_sam_sam · 2018-04-04 13:58

Between now and the weekend I will post the part numbers and photo of the board I still have to check something on it before I take it out of it case

The chip that controls the optic sensors I need pull that chip out and power it up again and see if the 100 watt light bulb in place of the fuse is still very bright or what or no light at all

sam_sam_sam · 2018-04-04 14:00

It has not damage traces on the board or any burn parts that I can see

frank freedman · 2018-04-05 06:01

Get pictures of the chip before powering up so that you have the numbers on the chip in case it decides to release the magic smoke.... or maybe pops its top.

sam_sam_sam · 2018-04-05 21:40

Thanks

frank freedman wrote: »

Get pictures of the chip before powering up so that you have the numbers on the chip in case it decides to release the magic smoke.... or maybe pops its top.

That sounds like a good idea

Remember I am using a 100 watt light bulb instead of a fuse

sam_sam_sam · 2018-04-05 23:53

Genetix wrote: »

Have you tried NTE for a replacement chip?

Do you have a link to where I can do a cross over replacement ( a good site )

frank freedman · 2018-04-06 04:37

As Genetix wrote, http://www.nteinc.com/ is a good start. You may have to search old datasheet sites though to find the specs of your current device and then hunt or a reasonable match. What type of chip is it? any readable/useful numbers on the case? What device is it out of?

sam_sam_sam · 2018-04-12 21:26

frank freedman wrote: »

Get pictures of the chip before powering up so that you have the numbers on the chip in case it decides to release the magic smoke.... or maybe pops its top.

I have decided to order the IC chips that the optic sensors are being told to turn on from
Here is a photo of the chips I bought

sam_sam_sam · 2018-04-12 21:30

Here is the photo of the IC chip it take two of them

jmg · 2018-04-12 21:33

sam_sam_sam wrote: »

Here is the photo of the IC chip it take two of them

Those are programmable Logic, so will not do anything until you get code into them.
Do you have the code ? Do you have a programmer ?

Can you work out what Logic might be inside them - do you have any known-good ones ?

sam_sam_sam · 2018-04-12 21:40

This is the problem I do not know if the ones that are in there are good or not it had a power supply surge of some sort

What makes me think this might be the problem it has all the optic sensors are turn on at the same power level and it if you look at the data sheet it shows that the power supply is from - voltage that turns it off and a + voltage turn it on

The optic sensors that it has ( HP then the part number )
Here is the data sheet for the optic sensors
https://www.promelec.ru/pdf/HCPL-3120-J312_HCNW3120.pdf

sam_sam_sam · 2018-04-12 21:47

jmg wrote: »

sam_sam_sam wrote: »

Here is the photo of the IC chip it take two of them

Those are programmable Logic, so will not do anything until you get code into them.
Do you have the code ? Do you have a programmer ?

Can you work out what Logic might be inside them - do you have any known-good ones ?

Do you have a programmer to do this

sam_sam_sam wrote: »

This is the problem I do not know if the ones that are in there are good or not it had a power supply surge of some sort

What makes me think this might be the problem it has all the optic sensors are turn on at the same power level and it if you look at the data sheet it shows that the power supply is from ( - voltage ) that turns it off and a ( + voltage ) turn it on

jmg · 2018-04-12 22:10

sam_sam_sam wrote: »

Do you have a programmer to do this

Most universal programmers will program PEEL18CV8, but you need to create the code first...
WinPLACE was what the ICT software was called, not sure where you download that these days...**
That software compiles boolean equations to JED file, which then goes into the PEEL18CV8.

You might get lucky, and be able to use ATF16V8, which is still an active part, and software is
https://www.microchip.com/design-centers/programmable-logic/spld-cpld/tools/software/wincupl

The 18CV8 is a little smarter than a 16V8, so not all designs can fit, but most decode-logic stuff can.

** addit : google finds some chatter and links here http://www.vcfed.org/forum/archive/index.php/t-40469.html

Peter Jakacki · 2018-04-13 03:09

I wrote a PLD compiler and programmer in Forth back in the day when we used such devices including ones with internal registered feedback paths such as GALs, ERASICs etc. In fact I wrote the compiler in one afternoon in one page of code. PLD source was just pasted into Forth and compiled using the PLD vocabulary. Here's a Forth'ish source snippet:

RAM !		= ( PH2I & RAMMAP & A15 ! )			\ 0000-7FFF
		# ( PH2I & ROMMAP & A15 ! & A14 ! & A13 ! )	\ 0000-1FFF
		# ( PH2I & SYSEML & A15 & A14 & IOCYCLE ! )	\ C000-FFFF
		# ( PH2I & SYSEML & A15 ! ) ;			\ 0000-7FFF

I had to reverse engineer the in-circuit programming and map but it saved me being dependent upon a $10,000 programmer and PC software!

Now back to the OP, as has been said, the chips are no good to you unless they are programmed, or you have the source/jed file and can program them. But I somehow doubt they are faulty. However as a generic test for any semiconductor component I use the diode test setting on the meter. Get any chip, perhaps a Prop and try it from Vss (ground) to and you will notice that similar pins such as I/O will read much the same. Then reverse the polarity of the probes and check again, they should all be much the same. You can even repeat this procedure from Vdd/Vcc (logic supply) as well. Most chips are damaged from ESD and this test shows up these faults quickly and easily.

Genetix · 2018-04-13 17:03

Peter,

I should try your diode test on a supposedly bad BS2.
Can I do this with any meter such as the Harbor Freight freebies?

I know that I have not been on this forum for a very long time life has gotten in the way

Comments