Problems before I've even really started (modemachines kit)

[twisted-space]

Hi all,
I bought a Xoxbox kit on ebay UK. I've started building it tonight and hit my first problem almost immediately.
I have the psu socket fitted, getting 10.8v AC at it's terminals fitted the diodes D40 through D43 and capacators C4 and C5. I've checked the diodes and capacators are correctly orientated. The build guide says that I should be getting ~15V dc across both C4 and C5, I'm getting 12.7 volts across C5 and 0v across C4.

I've done some hunting on the forum to try and find hints as what I should do and in the process discovered that I have a modemachines kit and that these are known to be poor quality. Not sure what to do now, I don't see that I can ask the seller for a refund, as he didn't mention what make the kit was and I stupidly didn't ask.

I'm a bit gutted to be honest, 7 components into the build and I'm already having problems.

Does anyone have any suggestions as to what I should do to troubleshoot this.

[adafruit_support_rick]

Sounds like maybe you have some cold solder joints? Make sure that all the leads are well-soldered. The solder joints should have a nice, smooth shiny flowing look to them. If they look blobby or crusty, re-solder them.

Post a picture if you're not sure.

[MMM]

If you have voltage across C5 but not C4 it points (in my opinion) to one of the C4 connections. Since C4 and C5 share one connection it would be a good idea to start there.

Please check (without power) continuity from the "-" terminal of C5 (that's the one closer to C4) to the terminal of C4 directly in the corner (there's a trace connecting C5 via one of the pins of the regulator to C4).

[twisted-space]

Sounds like maybe you have some cold solder joints? Make sure that all the leads are well-soldered. The solder joints should have a nice, smooth shiny flowing look to them. If they look blobby or crusty, re-solder them.

Post a picture if you're not sure.

Thanks driverblock, I'm pretty sure it's not a cold joint, I've soldered before (though not for quite a while) and I seem to have a good flow.

If you have voltage across C5 but not C4 it points (in my opinion) to one of the C4 connections. Since C4 and C5 share one connection it would be a good idea to start there.

Please check (without power) continuity from the "-" terminal of C5 (that's the one closer to C4) to the terminal of C4 directly in the corner (there's a trace connecting C5 via one of the pins of the regulator to C4).

My mistake, I wrote C4 instead C3, I have 0v across C3. The trace goes from C5 -ve through C3 +ve to one pin of the regulator then on to one side of C4, I have continuity all the way across. I haven't fitted C4 yet. I also checked I have continuity from C3 -ve to the other side of where C4 will go.
Apologies for any confusion.

[MMM]

My mistake, I wrote C4 instead C3, I have 0v across C3. The trace goes from C5 -ve through C3 +ve to one pin of the regulator then on to one side of C4, I have continuity all the way across. I haven't fitted C4 yet. I also checked I have continuity from C3 -ve to the other side of where C4 will go.

If you have continuity between C5 -ve and C3 +ve it probably means that there is something wrong on the other side (which actually is the ground plane). Do you have continuity of C3 -ve to the overall ground plane? If so, do you have continuity to D40 anode (the outermost diode)?

Is there a name on the PCB? There have been different types of IOboard PCBs; and it's important that we know which one we are talking about.

[twisted-space]

Do you have continuity of C3 -ve to the overall ground plane?

Yes

If so, do you have continuity to D40 anode (the outermost diode)?

Yes

Is there a name on the PCB? There have been different types of IOboard PCBs; and it's important that we know which one we are talking about.

Yep, it's MJB1ROLLIN.

[MMM]

Is there a name on the PCB? There have been different types of IOboard PCBs; and it's important that we know which one we are talking about.

Yep, it's MJB1ROLLIN.

Ok, I'll look into this tomorrow. Please check D40 and D41 in the meantime.

[twisted-space]

Ok, I'll look into this tomorrow. Please check D40 and D41 in the meantime.

Thanks MMM, I appreciate your help.

What do you mean by check D40 and D41? Remove and do a diode check? Or can they be tested in place?

[guest]

check your multimeter
and see if it has a symbol of a diode on it
this is the diode check mode
you put the probes across the diode one way
and it should read open circuit
and putting the probes the other way should give .6v or so

what powersupply are you using
is it an ic powersupply
is there a jumper for the powerswitch on the board
is that jumper shorted or the powerswitch installed

[twisted-space]

Thanks guest, yes my meter has a diode check, I thought that I would need to remove the diodes from the board to test them though. Will being in circuit not affect the test.

I'm using a no brand 9V AC/AC 500ma psu, it tests ok @ 10.7V. I've not fitted the switch yet and I'm not sure about the jumper, will check in the morning as it's gone midnight here, I'm heading to bed.

[guest]

ok
if there is no switch in place
then that explains why its not working

you can usually test diodes in circuit
just make sure the power is off

[twisted-space]

ok
if there is no switch in place
then that explains why its not working

That did it!! Thank you guest, you're a star.

Shame that the mode machines build guide is:

1 in german
2 shows a different board (looks the same as the adafruit one), with a different location for the power switch.

The english quick build guide shows my board, but specifically states that the power swith shouldn't be fitted until the board is fully assembled!

I suppose poor documentation is one of the reasons these kits have a bad reputation.

[MMM]

One problem with the kits is (or was) that different parts (most notably PCB's) were used that occasionally differed in some way (as you have already noticed). Another problem is that Mode Machines doesn't even have one single kit anymore (which makes it a bit difficult for me to debug things since I don't know the exact contents of the kits).

Which leads to another possible problem:

To save you from further frustration please check Q22 (the 2SC2291) and report the code printed on it.

[twisted-space]

One problem with the kits is (or was) that different parts (most notably PCB's) were used that occasionally differed in some way (as you have already noticed). Another problem is that Mode Machines doesn't even have one single kit anymore (which makes it a bit difficult for me to debug things since I don't know the exact contents of the kits).

Which leads to another possible problem:

To save you from further frustration please check Q22 (the 2SC2291) and report the code printed on it.

It's gone pretty well today to be fair. I have what appears to be a working I/Oboard, it powers up, the usb interface is recognised by my pc, I get 11.99V and 5.33V @ J4 (though I am at the top of the adjustment on TM6 to do so).


The code on Q22 is C2291 71G.

Thanks again.

[MMM]

The code on Q22 is C2291 71G.

This one should work. A number of the kits was shipped with a "fake" Q22 (probably relabelled).

You may also want to check if the kit contains tantalum caps for C42 and C62; some kits were shipped with ceramic caps instead.

Feel free to contact me (also by personal message) if anything else should come up with that kit.