Assembled TF536 - help needed

[sampedenawa]

Hello everybody !

I assembled my TF536, but I am having a number issues I was enable to spot correctly.

The board boots, but shows the red "expansion board diagnostic" defective screen (correctly showing manifacture 5080= Stephen, product=2).

If I continue, I can boot from floppy, load WB and run stable for hours, BUT:

- Do not see any fast mem

Additionally, IDE seems to be active, I can select "DH0" device from os 3.1/3.1.4 boot list, but if I try to load wb from IDE, it never comes to end (ide acrivity led continuously blinkin as if it loads from CF, but actually I don't think it does)

I triple checked all solderings with a scope, did not spot any shorts or missing solderings

Memory chips are MT48LC16M16A, so they should be the most compatible.


Would anyone post a close hires pic of the top side of the board I can use as reference ? Maybe I misplaced some IC (i don't think so, but anyway ....)

Anyone has some hint to troubleshoot ?

Thanks !

[PhilC]

I suspect you have more than one badly soldered pin on the cpld. Also if there is a problem with the ram buffers you'll get no ram detected.

Have you tried Diagrom?

Maybe post some photos of your board here and we will have a look.

[terriblefire]

You get this issue when the RAM is completely dead.

Have you booted DiagROM and looked at the address range 0x40000000 to 0x44000000?

[sampedenawa]

Thanks for your immediate feedback !

@Stephen
I ran diagrom, and it did not detect anything in the range you quoted. It only detect chipram and anything else.

Here alleged a few pics (hope the detail is enough).
Sorry for the board still not completely clened from flux.
Any further suggestion much appreciated.

[PhilC]

@sampedenawa in your photos its hard to tell but it looks like there are bridges across some of the cpld pins.

[PaulJ_2.0]

Flood with flux (too much) and reflow the ram and cpld.

[terriblefire]

@Stephen
I ran diagrom, and it did not detect anything in the range you quoted. It only detect chipram and anything else.

Did you *look* at it with the memory viewer?

[sampedenawa]

@sampedenawa in your photos its hard to tell but it looks like there are bridges across some of the cpld pins.

Flood with flux (too much) and reflow the ram and cpld.

Probably is a combination of flux remains and (too) low details of the pics, but I checked the pins of the cpld with a digital uscope, and I did not find any evident bridge.

Did you *look* at it with the memory viewer?

Here is the result:




Anyway, I will follow the advice form PaulJ_2.0, may some hidden bridge which could vanish by doing this.

Lateral thinking: any chance to have a counterfeit CPLD ? Could it be a plausible scenario ?

Will update later this evening.

Thank you all for now !

[terriblefire]

Well we can tell you exactly which pins are bridged if you do a complete memory detection and try editing the memory and see which byte lanes arent happy.

This is the 4th or 5th thread with exactly the same symptoms. all 4 were resolved using memory editor, complete memory detection and some magnification.

[terriblefire]

@sampedenawa please stop externally linking to photos. Its against the forum rules. if you do it again the thread will be deleted.