Acquired a IIfx (yay!) Issues (boo!)

[nyef]

Okay, the (apparently bomarc?) schematic that someone pointed me to says that pin 11 on the floppy connectors runs to the positive side of C196/C205 (one or the other), clearly decoupling, and also runs through L12 (an inductor, damps noise on the power rail) to the 5v supply.

So, check continuity from pin 11 to the capacitor and to the inductor. Check resistance across the capacitor (should be high, preferably reading as open circuit) and the inductor (should be low). Check continuity from the inductor to the 5v intake at the PSU. Check continuity from the capacitor to the ground pins on the floppy connectors. Look for signs of leaked electrolyte or other potential corrosives. Look for sign of damaged PCB traces in the area.

 

[nyef]

While I'm thinking about it, one other check, this one while the machine is powered up: Check the voltage levels relative to ground on both sides of the inductor.

 

[mst3k]

Okay, the (apparently bomarc?) schematic that someone pointed me to says that pin 11 on the floppy connectors runs to the positive side of C196/C205 (one or the other), clearly decoupling, and also runs through L12 (an inductor, damps noise on the power rail) to the 5v supply.

So, check continuity from pin 11 to the capacitor and to the inductor. Check resistance across the capacitor (should be high, preferably reading as open circuit) and the inductor (should be low). Check continuity from the inductor to the 5v intake at the PSU. Check continuity from the capacitor to the ground pins on the floppy connectors. Look for signs of leaked electrolyte or other potential corrosives. Look for sign of damaged PCB traces in the area.

OK.
Continuity good from pin 11 to c196/c205 on both FDD connectors.
No continuity from pin 11 to L12 on both FDD connectors.
No continuity from 5v rail to L12, but there is continuity from 5v rail to LC11 (for whatever that's worth).
Voltage, with power on, at L12 is 0.

I see no signs of broken traces or weirdness around L12, but I'll put it under the digital microscope this afternoon.

Faulty inductor at L12? There is a continuity signal thru L12 (from pad to pad) but I'm guessing that means nada.
Thanks yet again. I really do appreciate the guidance.

 

[mst3k]

While I'm thinking about it, one other check, this one while the machine is powered up: Check the voltage levels relative to ground on both sides of the inductor.

Double checked this (L12) and get 1.3v on either side.
L11 has 12v on either side.

 

[mst3k]

Greetings,

Grabbed the Bomarc schematics and stared at them for about 15 minutes.
It looked to me like the key to this problem, and the only place I hadn't checked was the inductor at L13.
Checked and it showed 5v.
Put it under the scope and found this:



I found a broken trace that was folded right along the yellow print line.
It runs under the connector and I'm not sure how to effect a permanent repair.
I soldered a bodge wire to the inductor pad and wedged it under the connector to the base of pin 11.
Fired up the IIfx and checked voltages. Voila. 5v on pin 11. Powered down and attached a FDD.
Buzz, whir, it worked. Worked, that is, until the bodge wire came loose.
How am I going to solder underneath the connector (slaps forehead)? I hate to take the connector off but not sure what other option I have...
Anyhow. It looks like I've solved one problem. On to the next!
Thanks to nyef for the hand holding.

 

[obsolete]

How am I going to solder underneath the connector (slaps forehead)? I hate to take the connector off but not sure what other option I have...

I don't have any experience with the IIfx specifically, but my experience with that style of connector in general is that the plastic shroud is just friction-fit over the metal pins, and it's possible to just pull the shroud off, perhaps assisted by some heat.

If it was my machine, though, I think I'd just solder a bodge wire a little more securely rather than trying to effect a repair under the connector.

 

[Callan]

You can desoldrr the header pins, or... if you don't care about looks you can just run a jumper from the inductor from the top of the board to the bottom of the board. It would be a long run, but easy to do.

 

[danny.gonzalez.0861@gmai]

I feel your pain on this one, I try to make my repairs look nice but I too struggle to find a nice solution for you. Maybe if you run a bodge around the connector and use the trace in between the connectors to join them. I made a rough drawing so you can see what I am talking about: (red line)

You will need a fine tip soldering iron and a steady hand but I think its doable.

 

[mst3k]

I decided to just run a temporary bodge from L13 through the nearby mounting hole and back around to pin 11 underneath.
I'm not sure what the permanent fix will be, but for now I just wanted to get the drive working.
Well, the drive now gets power (5.1v), but the drives ( and I've tried three including my Floppy Emu) are not recognized.
When the computer starts the FDD's do their seek thing (the Floppy Emu powers up and self tests) however the machine boots without ever seeing the drive(s) and goes to the desktop (booting off a ZuluSCSI or MacSD with 7.5.3 or 7.1.1).
As before, everything else works, but no matter which FDD or cable I use, the FDD's are not recognized.
So, the power problem is solved, but where I should I look to see why they aren't be seen?
I've checked the continuity to the SWIM chip (as per nyef) and there is continuity to UC10 on the even numbered pins.
Could the SWIM chip be bad? Is there somewhere else I should be looking?
Would throwing the whole thing across the room help?
Thanks

 

[Callan]

I'm wondering if someone cut that trace on purpose. Hmm
Scope time.
Verify each of your pins are correct (sounds like you've done that). Verify all voltage and gnd pins are good. Verify your enable line is right (active low) and that read data has activity.same with all the other lines After that I would trace out everything back to the swim. I would use a scope to verify each of the lines for activity and verify the signal @ the corresponding swim pin. If all that checked out I'd suspect the swim. I'm definetly not an expert on this board, but that would be my first step. Then I'd do what your doing and ask the experts.on here.

 

[mst3k]

I'm wondering if someone cut that trace on purpose. Hmm
Scope time.
Verify each of your pins are correct (sounds like you've done that). Verify all voltage and gnd pins are good. Verify your enable line is right (active low) and that read data has activity.same with all the other lines After that I would trace out everything back to the swim. I would use a scope to verify each of the lines for activity and verify the signal @ the corresponding swim pin. If all that checked out I'd suspect the swim. I'm definetly not an expert on this board, but that would be my first step. Then I'd do what your doing and ask the experts.on here.

Thanks Callan.
I don't have a scope, but I just bought one (a cheapie handheld that Adrian Black demo'd on his YouTube channel) and I'm looking forward to testing the board.
It's interesting how one gets 'target fixation' on a non-working aspect of these old machines. Realistically, I can use the IIfx now without the FDD and could probably sell it for what I paid for it, but the urge to fix that which is not working is overwhelming.
Thanks again. I'll post results when my learning curve for using a scope has leveled out enough for me to report.
Cheers.

 

[Callan]

You could use a logic probe to verify activity. You just want to make sure your pins are getting data from the swim. If your in my area I'd be happy to help. In the interim you could just build your drive images on another machin, but i would really miss my floppy emu.
And... since you mentioned it. If you decide to pass it on I'd be interested since I don't have one.

 

[mst3k]

You could use a logic probe to verify activity. You just want to make sure your pins are getting data from the swim. If your in my area I'd be happy to help. In the interim you could just build your drive images on another machin, but i would really miss my floppy emu.
And... since you mentioned it. If you decide to pass it on I'd be interested since I don't have one.

I'll keep you in mind if me and the IIfx have a falling out, but despite my frustrations, I do intend to keep this beast. It took a while to find one I could afford!
My comment was more about how repairing it becomes as important as owning it. And your observation about the Floppy Emu is spot on.
Hard to live without that bugger.

 

[mst3k]

You could use a logic probe to verify activity. You just want to make sure your pins are getting data from the swim. If your in my area I'd be happy to help. In the interim you could just build your drive images on another machin, but i would really miss my floppy emu.
And... since you mentioned it. If you decide to pass it on I'd be interested since I don't have one.

OK. I put my junior electronics explorer hat on and dove in..
On the FDD connector the odd numbered pins all show the correct voltages as per your pin diagram.
On the FDD connector the even numbered pins all show continuity to the SWIM chip at UC10 (except PIN 20)
Using the junior electronics explorer $25 pocket oscilloscope, I get beautiful wave forms from PINs 16,14,12. I get jittery, low wave forms for Pins 10-2 but consistent. Nothing on Pin 20 and a regular scramble fest on pin 18. It jumps around at first and then just jitters.
Other than assuming that PINS 12, 14 and 16 are behaving correctly, I do not know what this means.

Okay, next phase is to check the internal floppy connectors, J16 and J17. Pins 14 and 20 shouldn't be connected to each other (20 is a no-connect generally, and 14 is drive select), but the others should show continuity (though you might have trouble measuring that for the odd-numbered pins, those are all power and/or ground).

I only show continuity between the odd numbered pins in the connectors.
There is no continuity between the even numbered pins and the odd numbered pins, or between the even numbered pins.
Pins 20 and 14 do not show continuity, but neither does any of the even numbered pins (if I'm understanding what nyef is saying).

So, any ideas?
Thanks