Real Bull Controller box question

[sacentre]

Good advice there from Homebrewed. I salute your courage in diving into this thing- there's a certain satisfaction in fixike you ale ling something with little info to go on.
Even if the IC number has been erased, you might be able to locate it online by typing in the function "3-1/2 digit tachometer IC" or something like that.
appears, I'm less inlined to
The hunting you mentioned could be a factor or not. I wouldn't expect the brushes to be worn unless the motor has lots of hours on it.
If you can read the part # from the opto interrupter post it here- I have some opto databooks

It's people like yourself and Homebrew and your willingness to help that deserve the credit. I'm not very courageous - only desperate! After running the test with the drill over and over, I'm still not seeing the fault. I can easily try warming the board a bit with a hair dryer as Homebrew suggests. Until and unless it appears under the test conditions, I'm less inclined to see this as a fault on the display/encoder. As for brushes, I'll be checking those later today. Fortunately, I have both a spare set as well as a new, unused motor.

What's worrying me now is the possibility of a fault of some kind with the KB controller itself. I never asked Amadeal what warranty if any, came with the box. I will have to talk to them about returns if it turns out to be faulty. I haven't the knowledge or the tools to establish that other than by the process of elimination, though. Still, I'm not not at that stage yet.
Thanks again for the continuing help.
Trevor

 

[homebrewed]

If you feel up to it, pulling the tach board and looking for cold solder joints might pay off. If nothing else, unplugging/reconnecting the connectors might clear up a contact problem. If the IC is socketed, lift it up a bit and push it down again -- same reason.

 

[sacentre]

If you feel up to it, pulling the tach board and looking for cold solder joints might pay off. If nothing else, unplugging/reconnecting the connectors might clear up a contact problem. If the IC is socketed, lift it up a bit and push it down again -- same reason.

Yes, all good suggestions and easy to carry out. I'll do that and check that IC ID at the same time.
Thanks
Trevor

 

[sacentre]

Just as I suspected, the cynical manufacturers have scratched off the ID.

 

[sacentre]

Ok, I reported yesterday that I couldn't get the display to go faulty using the B&D method. Today, I have the display PCB out on the bench and ran the test again this time heating the PCB with a hair dryer. It was too hot too touch at one point. Still nothing.

On the next go (no hair dryer) the fault appeared (see video). It's possible I didn't run it long enough. Now, the behavior is pretty consistent and repeatable - flashing 8888 after 2 or 3 minutes without adding heat. So, it seems fair to assume it's not a faulty controller board or motor brushes.

There's still the possibility that there's some problem with the DC feed from the filter board OR the opto-isolator itself. I will need to figure out the routing on the PCB so I can duplicate an external DC feed.

 

[markba633csi]

OK. It's quite possible that it could be a faulty component. It's also possible that the way the factory put the unit together the controller was able to transfer a voltage surge into the readout board; prone to fail by design. This may be an endemic problem with all of these- who knows
Maybe it's time to consider either replacing the whole unit or getting a spare readout board since you don't have the means to troubleshoot down to component level
The readout chip is probably CMOS and those can be damaged by static discharge, voltage surges, lightning and even solar flares.
And the environment it's in next to the controller is definitely a noisy one, electrically speaking but the chip could have been damaged just by handling at the factory
It probably will fail even with the spindle stationary but without test equipment or spare parts to substitute it's hard to tell which part is failing
I don't think you need to bother with the external dc test, but it can't hurt to try. A little 9 volt battery would probably work

 

[sacentre]

OK. It's quite possible that it could be a faulty component.
Maybe it's time to consider either replacing the whole unit or getting a spare readout board since you don't have the means to troubleshoot down to component level
The readout chip is probably CMOS and those can be damaged by static discharge, voltage surges and the phase of the moon, among other things.
And the environment it's in next to the controller is definitely a noisy one, electrically speaking, but the chip could have been damaged just by handling at the factory
It probably will fail even with the spindle stationary but without test equipment or spare parts to substitute it's hard to tell which part is failing

Yes, I was almost at the stage where I was ready to just order a replacement.

I might still try running the module on an external DC supply which should be within my capabilities. If the fault is still happening that will wrap it up as far as a faulty module is concerned. I'll report back whatever the final outcome is. Meanwhile, many thanks once again to you and Homebrewed for helping me out with this.

 

[homebrewed]

Yes, I was almost at the stage where I was ready to just order a replacement.

I might still try running the module on an external DC supply which should be within my capabilities. If the fault is still happening that will wrap it up as far as a faulty module is concerned. I'll report back whatever the final outcome is. Meanwhile, many thanks once again to you and Homebrewed for helping me out with this.

Grizzly stocks the display board. $60USD not counting shipping. You may be able to do better elsewhere since you're in Singapore. Naturally, they don't have a product manual so it's impossible to say if the flashing display is an indication of an error condition or not.

Grizzly's product page shows a photo of the part, and I was able to get the chip P/N from that. It's a Holtek HT48R30A-1, which is a RISC microprocessor chip. It's got a 2K program space and 96 bits of RAM so your average Arduino stomps on it.

The photo doesn't look exactly like yours but I'm pretty sure it is a drop-in replacement. However, before you order a replacement I'd highly recommend measuring the power supply voltage while it's misbehaving. It would suck if it turns out the problem is a bad controller board.

If your external supply is variable it would be interesting to lower the voltage to see if the display starts flashing. That also would be a good indication that it's not the display board (but not good for your wallet, unless the controller is still under warranty).

 

[sacentre]

Thanks for the info, Homebrewed. Especially the IC P/N. I'll check check out.

$60USD? I was hesitating to drop 18GBP plus shipping on a replacement but that price seems reasonable in comparison. Still, if I knew for sure that's all I need to pay to get rid of this problem, I'd order one immediately.

All this stuff is knocked out by the million in factories in China probably at a couple of dollars a piece if that. By the time they're shipped to Europe or the US, marked up and re-exported to me, they're selling somewhere around the price of gold dust. Basic economics, of course. It's a pity that being in the Far East already, I can't source them nearer. Even if I knew the OEM, they'd probably have a minimum order of 100PCS.

I still have a way to go to establish exactly which part of the system is faulty but I think we've narrowed this down to the display module itself, the encoder or the filter board which feeds the DC to the display. All these are available from Amadeal as spares.

As discussed, all that remains to try now is getting the encoder and display PCB wired up to bench PSU and rigging up a makeshift rotating disk and seeing if I can see the fault. If not, this will suggest the filter board (possibly a noisy DC line?) I'm also inclined to think that the flashing 8888 is a designed error signal but what the error is, who knows?

Once I can say with any certainty which part is faulty, I'll approach Amadeal to see if I can negotiate some kind of return/replacement but I may wait till after Christmas now before doing that.

Thanks again for the input.

 

[homebrewed]

Thanks for the info, Homebrewed. $60USD? I was hesitating to drop 18GBP plus shipping on a replacement but that price seems reasonable in comparison.

All this stuff is knocked out by the million in factories in China probably at a couple of dollars a piece if that. By the time they're shipped to Europe or the US, marked up and re-exported to me, they're selling somewhere around the price of gold dust. Basic economics, of course. It's a pity that being in the Far East already, I can't source them nearer. Even if I knew the OEM, they'd probably have a minimum order of 100PCS.

I still have a way to go to establish exactly which part of the system is faulty but have narrowed this down to the display module itself, the encoder or the filter board which feeds the DC to the display. All these are available from Amadeal as spares. As discussed, all that remains to try now is getting the encoder and display PCB wired up on my bench to an external DV supply and rigging up a makeshift rotating disk and seeing if I can see the fault. If not, this will suggest the filter board.

Once I can say with any certainty which part is faulty, I'll approach Amadeal to see if I can negotiate some kind of return/replacement but I may wait till after Christmas now before I do that.

Thanks again for the input.

Modern-day supply chain stuff can be strange, that's for sure. In your case, it sounds like a case of so near but so far.

If you have a signal generator you don't need to make a rotating disk setup in order to exercise the board. It wouldn't have to be much -- even a 555 timer could emulate that interrupter/detector combo. That's easy for me to say since I have all kinds of electronic components on hand, but it's worth mentioning. Over here, you can buy a 555 timer for under a buck. It also is easy to make a square wave oscillator using either CMOS inverters, NOR or NAND gates.

BTW, looking at the display board it looks to me like the interrupter signal probably is routed through an opto isolator (it's the 4-pin IC on the board). So you can't assume that one of the two interrupter inputs is the same ground as the board's power Gnd input. I don't think that means you have to have two power supplies to test the display board. The interrupter input's negative signal line could be connected to board Gnd.