Short reply to another question. The box has been powered up for a few hours now. The CPU heatsink isn’t stone cold, but I can’t see that it’s any warmer than the chassis. The heatsinks on the PSU are only very slightly warm, but at least the kit below them is doing something, whereas I’m guessing the CPU has nothing to do at the moment.
Back to the serial voltage. We established my male RS232 port on the PC was delivering 11.2v to the box, but the USB is limited to 5v. Does that matter?
I do not think the blobs are anything mysterious, the 2 devices are programmable and would have been programmed before being fitted to the board. The marks I think are 2 show that they have been programmed and the blue mark on the flash device (U12) is probably to show that it has been verified - double checked to avoid having to remove it during production. The device in the socket (U17) can be removed if it has a problem the flash device would have to be unsoldered to replace it.
I am struggling to find data on the devices but will keep trying. I would really like to know what voltage is on CB67 small capacitor near the flash device U12, the one end will be 0V and the other (I think) is the supply for U12. The voltages on U3 (above the CR2032 cell) may also be interesting (middle pin is 0V). To be continued...
OK, will check tomorrow, but…
Oops! Didn’t notice the last para to Stroker’s post of 20/02.
The bit about the component to the right of the heatsink I think I answered inadvertently because I wondered what the blob was.
Now big resistors? Are they the handsome green things with stripes labelled R200 and R94 – hey what’s that doing here, R93, R95 and R96 are old friends from the PSU board. Zero ohms – both multimeter. But then the green one on the PSU boars, R93, also shows zero ohms. Is that close enough to 0.5 ohms?
Sorry I did not notice your earlier post - It would seem to me that the CPU is not running but this is no real surprise since we have forced the main power on with the link on the PSU so it has not powered up in the normal controlled manner. Having said that I would assume the CPU should also be running in standby but perhaps using a low power mode. The CPU needs a program to run (in the Flash device), memory (the 4 Hynix devices) and power supplies to these devices. The CPU also needs a clock and a power-on-reset signal to kick it off. So the 64 million dollar question is what are we lacking and why? I need to do some thinking - I notice there are some test points near the CPU marked as TP followed by a number but we don't know what each is meant to show us - life's a bitch!
Just had a thought perhaps you could replace the link on the PSU with a switch, then we could power up with just the standby supplies and then close the switch to bring on the main supplies. If you do this be aware that high voltages will be on the switch - any switch will do - perhaps a light switch. But we really need to work out how the standby switch is supposed to bring on the main supplies - perhaps the switch idea would just be a waste of time.
Concerning voltages USB uses levels of 0V and 5V, RS232 levels of +12V and -12V so although they are both serial interfaces they use different voltage levels and are not directly compatible.
R200 and R94 was expecting low values so that sounds about right - a stab in the dark really. Component numbers usually only relate to the board they are on, so you could have for instance R5 on the power supply another R5 on the motherboard and yet another on the display board and each of these could be different values and be in no way be related or connected.
I think I need a drink!
Right, perhaps you could check this out first, The memory runs on 2.5V not sure where that comes from just yet and U3 (NE57811) produces some related signals including a 1.25V reference. The 4 Hynix memory devices have the supplies on the corner pins with one side's corner pins being 0V and the other 2.5V
U3 has 5 pins if you look at it with the pins at the bottom and the body at the top the pins are numbered left to right 1 to 5 with the middle pin being 0V. Pin1 and 2 should be 1.25V and 2.5V, pin5 should also be 1.25V, pin4 is a called shutdown, not sure what to expect on this. Again be careful not short pins with the meter probes.
I think this is a good place to start and will hopefully give us a clue as to is or isn't happening.
Please check for 2.5V on each of the 4 Memory devices - I think they may not all be supplied from the same source.
I'd like a pic of devices below the JTAG connector and to the left of the CPU ie from U23 to U3 include the 2 memories - thanks.
Pin 1 1.31v
Pin 4 2.62v
Pin 5 1.31v
Hynix memory chips: 2.62v on all four.
I did find it very difficult to measure Pins 2 and 4 – several times got nothing, and several times killed off some voltages completely – HDD stopped clicking and failing fan stopped grinding. So tried to heed your warnings but a probe obviously slipped – maybe I shorted 1 and 2 for instance. When that happened +5VS ws still present.
Hope the pic is OK.
Pic is good thanks, voltages seem reasonable, not sure where we go from here. Worried about probes shorting could do with finer probes.
Returning to think mode....
Sorry for the pause in posting I'm still searching, mainly in vain for data on some of the chips on the motherboard.
I have found data for U21 STV6412BD which is just used for switching video and audio inputs and outputs so we can probably rule this out as causing our problems.
The SM8952AC device in the socket near the battery is quite a complicated CPU/Controller/IO/Flash device which is much more interesting as regards our faults but I am still looking into this and may take some time to digest.
We have discovered 8 fuses (names starting with FB) on the motherboard next to the main power connector but unfortunately all are intact.
List of components next to connector below, courtesy of jphilweybr.
Between the caps and the PSU connector, there are lots of tiny components. Running from Pin 1 towards the back, they are:
I will not be posting quite as often over next couple of weeks due to other commitments but I will still looking at this when time allows.
I am taking a break from looking at the motherboard for the time being and concentrating on trying to get the display to work instead as this could give us some badly needed clues as to what is actually happening. Some of your recent voltage measurements suggest to me that there may be display data getting to the front panel board.
I am trying to convince myself that due to the loss of the 5VS supply the two transistors driving the transformer Q3 and Q4 could have blown.
There is a simple check on transistors you can do with a meter where the transistor can be tested as being two diodes with the common pin being the base. So you should see a diode between base and collector and another between base and emitter. If you look at the Q3/Q4 with the two pins at the bottom I think the base is the left hand pin of the two, so check for diodes from this pin to each of the other two pins.
If you don't follow this and you are feeling brave you could just go ahead and change Q3 and Q4 which I think you may have replacements for. If my theory is correct and the transistors you have are equivalent to the old ones I am hoping the display will then work - fingers crossed!
Many thanks for your continuing help Stroker. And a good time for you to take a break, because I took a nice holiday and found lots of things to sort out when I got home. One involving an Axxr laptop merits a thread in its own right but probably not on this site!
Anyway, I’ve managed a little bit of time today and here’s a brief report.
I’ve had a look at Q3 and Q4. I’ll call the pins Left, Right and Top. The following results were identical for both Q3 and Q4.
1. There is continuity to GND to the Right hand pin
2. The diode test was normal for the bottom two - + 6.5v with the red probe on the left and Infinity with it right.
3. Similarly, Red probe Left, Black probe upper +6.5v, infinity the other way around.
4. With red probe on the Right and black probe upper about 0.12v and infinity the other way around
5. There is no voltage between the chassis and the bottom pins. There is 5.07 to the top pins – the same as at the transformer middle pins on the back.
6. I did buy two of the BCW65 transistors, when they were a mere £1.21 each, but at the moment cannot find them.
I don’t know about D9. There is a blob on the L31 side, and I can measure 4.6v to it. However, while the pics show something at the other end, I get no voltage there. Nevertheless I can do a diode check on it. With the red probe on the L31 end, I get 6.5v and infinity the other way. I realise that I didn’t correctly identify D9 in my post of about three weeks ago, I was referring to a larger component – probably the label Y5 refers to it.
I think this is the best pic.
From your results I would say that Q3 and Q4 on the front panel are OK.
Your results for D9 on the motherboard are much more interesting. From your pics it looks pretty certain that the bottom end of D9 is connected to pin 10 on the Syncmos chip which according to the datasheet I downloaded is the reset pin. If I understand you correctly the bottom end of D9 is at 0V and is holding the chip in a reset state which would prevent it from doing anything!
I would expect the reset pin on the chip to be connected to a resistor and capacitor. D9 is there to discharge the capacitor quickly when power is turned off. So the resistor could be open circuit, the capacitor could be short circuit or there could be an internal short in the chip. There is also a possibility that there is something else connected to this circuit which is causing a short.
First thing to do is check for a short from bottom end of D9 to GND, if there is I would first suspect the capacitor next to D9 unfortunately to check this the motherboard will need to come out (as I am sure you already guessed). Check for a short across the cap to confirm we have the right cap, you will then have to remove the cap to check it is shorted and also check the short has gone from the board itself. If the cap is not shorted and the short is still present on the board take a pic of the underside of the board in this area so we can try to see if anything else is connected. If the cap is shorted replace it with the nearest value you have as long as the voltage rating is greater than 5V.
If there is not a short we need to check the resistor is intact and soldered properly. One end of the resistor would be connected to 5VS and the other to the capacitor and the bottom end of D9. I think the resistor is R228 on the other side of the capacitor - first try to read the details from the resistor, expect 3 digits - the first 2 digits are just digits the third is the number of noughts to stick on the end, for example 103 is 10K (1,0 and 3 more noughts). See if you can measure the resistor and check continuity, one end should be connected to the top end of D9 the other to the bottom end - you should be able to measure the resistor with your probes across D9 using a higher range (you should measure the resistor with the probes one way, and see D9 with them reversed).
I think the Syncmos chip, which includes a CPU, does all the low level and standby tasks, while the main CPU only operates when out of standby. The Syncmos chip presumable drives the front panel display and takes in signals from amongst other things the front panel push buttons. If I am correct on this the box will not do anything until we get the chip out of it's reset state. This could well be a huge breakthrough I just hope the cause is something easily fixed.
If Q3 and Q4 are OK, could there be something wrong with the transformer? Is there a way of checking that?
Resistance from chassis to the bottom end of D9 is 7.6kΩ. To the other end, I get OL.
Just to double-check, the next checks I have to do need access to the underside of the motherboard? I think R228 is 822 – see pic, so 8.2 kΩ?
I tried to post an example diagram of a POR circuit but was unable to upload it as usual!
I have sent it to you by email - perhaps you can post it on the thread for me - thanks.
Do not take the motherboard out just yet - you may not need to.
Since you measure 7k6 from bottom end of D9 to GND we do not have a short.
Try to measure R228 it should be 8k2 as you say, also measure voltage at each end - the cap should charge up to the same level as 5VS so you should have about 5V at both ends of R228. Because you measure 7k6 to GND the cap may not charge up enough.
I need you to confirm these measurements (R228 and voltage each end) before I ask you to do anything else. The cap may well be faulty but I need to be sure of the details - can you see what value the cap is? do you have something similar? I think the cap is between R228 and D9 - looks like CE47.
I would expect the transformer on the front panel to be OK since you have 5V on the top pins of Q3 and Q4. The 5V would be fed through the transformer primary windings.
R228 is tiny and not easy to get probes to make contact, but I finally managed to get a resistance reading of 7.7 kΩ. I didn’t managed to get a voltage though. I think I was making contact at least at one end, because the reading settled at 12.6mV, whereas without any contacts the meter usually runs back to zero.
CE47 is 50V 10µF. The one next to it is 16V 100µF. I have some 50V 22µF and some 10V 470µF.
Sorry, I can't upload that diagram either.
Yep those little resistors are a real pain.
Can you measure the voltage on the bottom and of D9? We need this to be over 3.5V
I think we need to bite the bullet and get the motherboard out, remove CE47 and test it, which I hope is faulty.
Also measure the resistance from bottom end of D9 to GND which I hope will be much higher than 7k6.
I would fit 16V 100uF test it first so we know it's good then put it all back together and see what happens.
I have high hopes for this - good luck!
Having internet probs.
If CE47 is not faulty check the cap next to it - I may have picked the wrong one.
Well with the motherboard still on, the resistance to the bottom end of D9 is 7.7 kΩ, 15Ω to the L31end (not OL as reported yesterday). There is just 12mV to the bottom end, 4.7v to the L31 end.
So I’ll get the motherboard off. It promises to be fiddly to get all the ports at the back loosened and removed. To double check, the cap I’m removing is 50V 10µF, so would my nearest replacement be some 50V 22µF?
I’ve managed to make a picture, albeit third rate, of your diagram, hopefully now uploaded. And if it’s the other one that’s faulty, would my 10V 470µF be OK?
Hang on a minute
You've only got 12mV on the bottom end of D9 - it's looking like R228 is either faulty or not soldered properly. Check this first before taking the motherboard out. Beware the resistor may come loose when you heat it and come off the board.
Perhaps a better thing to try for now - short D9 top to bottom with a pair of tweezers or pliers and see if anything happens. That will raise the reset pin to 5V and get it out of it's reset state. It will return to reset when you remove the short.