Voodoo 5 5500 & 4 4500

Position

• X-Tals Y1 / Y2
• TP2 / TP6
• R41 / R42 / R90 / R99
• Power Supply Gate Driving
• Main Power Supply Output
• Supplemental Power Supply
• –
• –
• –
• U13
• U4
• FET
• U12

Reading

• 14.318MHz
• 67,03MHz
• 166MHz (depends)
• 30% Duty Cycle @ ~300KHz
• 2.8V – 3.0V
• 3.2V – 3.3V
• –
• –
• –
• CAHCT1G125… / SN74AHCT1G125…
• TPS3801J125DCKR
• IRF8712
• – – –

Possible causes

• Power Connector not connected.
• Power Suppy not working.
• Master GPU defective.
• Bios chip out of order

Solution

• Connect the connector.
• Repair the Power supply.
• Replace master GPU.
• Replace Bios chip

Possible causes

• Power supply completely fucked up due to overaged capacitors and field effect transistors.
• –
• –

Solution

• Switching converter control chip must be replaced.
• All FET’s and rectifier diodes have to be replaced.
• All electrolytic and polymer capacitors have to be replaced.

Possible causes

• Do check the output voltage of power supply for correct value.
• Check if all FET’s are receiving signals on their gates by controller chip.
• Check if all FET’s on the positive rail getting their power.
• Measure, if the controller chip gets the start signal from AGP port.
• Look for a short circuit on GPU power rail.
• Check if something runs hot.

Solution

• If you find a fault, perform rectification of it an test again.
• If all is ok – replace master GPU
• –
• –
• –
• –

Possible causes

• One of the eight memory chips has died. Preferred some of them, that are tied to the master GPU.
• Some of the TSOP memories has a bad contact.

Solution

• Replace the faulty memory chip.
• Carry out contact pin reflow.
• –

Possible causes

• Wrong or experimantal Bios
• Bios corrupted
• Bios Chip nuked.

Solution

• Reflash the Card-Bios, using a second PCI graphics card and setup boot from PCI first.
• Replace Bios EEprom and flash it.

Possible causes

• Memory fault

Solution

• In the worst case change all memory chips.

Possible causes

• Bad Capacitors, particular check the seconrary polymer types, testing for leakage current, self-inductance and component quality.

Solution

• Replace the bad capacitors to the correct types. Important:
  Important to note that not every capacitor is suitable due to its key data and size. It is essential to ensure that the correct specification is used!

Possible causes

• Most likely bad solder joints related to the memory chips

Solution

• Check with magnifier (recommended) and solder it gently.

Possible causes

• This is an unpleasant defect, because there can be several reasons for it. First of all you should check the power supply of the VSA100 chips. Test if the power supply still works on both channels. Check the ripple of the DC voltage on the secondary side of the power supply. This can only be determined with an oscilloscope and also please check, if the voltage on the primary side is ok.
• Check if the two coils have a clean magnetic field spectrum. This can be seen very well with a spectrum analyzer and a HF-Probe.
• Check if something ist running too hot.
• Bad connections or traces.

Solution

• Fix secondary power supply.
• Try another or replace PC PSU.
• Correct interruptions of traces.
• Replace TSOP RAM Chips

Possible causes

• Memory for seconrary GPU failed.
• Secondary GPU failed.

Solution

• Repair that memory, seconrary GPU is tied to.
• Replace secondary GPU.

Possible causes

• This shows a memory adressing fault.

Solution

• Adressing line, or may be sync line could be broken and shold be repaired.
• May be a memory fault as well…
• May by a defective VSA100 (if occurs at post screen)

Possible causes

• This is actually a standard error of the TSOP RAM of the master GPU and occurs relatively often.

Solution

• Replace master GPU’s TSOP RAM and you will be a happy camper..

Possible causes

• Occurs particularly often with pci cards. This is usually because the GPU power is running up later than the memory power.

Solution

• Check if GPU Power ist present. If not repair the GPU Power Supply.
• Check whether the FETs are controlled at their gates.
• Replacing SC1102CS may help.

Possible causes

• The 2,8V Voltage hungs at ~1V and doesen’t come up or come up seconds delayed, so too late zu initialize the GPU.

Solution

• It is strongly recommended to replace all FETs and the Controller SC1175 in all due cases!

Possible causes

• U13 is bitchy to temperature or if card doesen’t starts not at all, C13 is nuked.

Solution

• Anyway – please replace U13! This fucking thing can blow up the whole card.

Possible causes

• You fucked up the solder joints
• After soldering heat the overaged TSOPs are killed.

Solution

• Please resolder the pins carefully.
• Replace the TSOPs.

Possible causes

• Secondary GPU inhibited – happened after gambling around with fancy heatsink stuff? For sure you have damaged something.

Solution

• Look for some damage you inflicted! May be R97 damaged – you should replace it. Keep your eyes peeled while working.

Possible causes

• If Memory Chips are to be new, GPU may fail.

Solution

• Replace Secondary GPU

Possible causes

• Serious GPU syncron and RAM timing failure

Solution

• Please replace all memory chips

Possible causes

• Memory fault, primary or secondary possible

Solution

• replace memory chips

Possible causes

• This may happen if both GPU’s are nearly out of sync.
• The secondardy GPU incorrectly synchronises on the falling edge due to unstable system clock.

Solution

• Please be sure to check the crystals for exact frequency. Replace if necessary.

Possible causes

• Overvoltage
• GPU fault

Solution

• Check voltage and adjust to 2,85V
• Replace respective GPU

Possible causes

• Primary VSA100 fault
• Card badly contaminated with silver containing heat compound
• Voltage fluctuating

Solution

• Replace Primry GPU
• Remove all chips and deep clean the device
• Loof for shorts and fault currents

Possible causes

• Card may be contaminated. (experience)
• Memory fault.

Solution

• Carry out professional cleaning with according agents.
• Replace master GPU’s memory chips.

Possible causes

• Memory fault as well

Solution

• Replace memory chips

Possible causes

• Most likely the BIOS Chip.
• May be the secondary memories. Sounds strange, but it is true.

Solution

• Please replace the BIOS EEPROM
• Replace all of the secondary RAM chips.

Possible causes

• RAM Chip

Solution

• If in Single Chip Mode – primary side. If in fast performance or FSAA only, secondary side. Replace TSOP RAM.

Possible causes

• RAM
• Bios EEprom

Solution

• Replace Primary RAM
• Replace Bios Chip. (That’s what helped here in this case)

Possible causes

• RAM

Solution

• If at bootscreen – replace primary RAM chips.

Possible causes

• Too low voltage for the RAMs

Solution

• Check the voltage for the RAM chips. It should have 3.3V. If not, repair power source.The voltage level in this case was to be found at a 300mV rippled 2.7VDC. Replacing the OP-Amp [AR1] and it’s FET [Q5] solved the problem.

Possible causes

• GPU (Master) overheats
• Insufficient cooling
• GPU defect
• BGA defect

Solution

• Ensure proper cooling
• Replace GPU
• –
• –

Possible causes

• Nuked BIOS Chip on the 3dfx card
• The flash attempt chokes out the mainboard bios.

Solution

• Replace and flash Bios Chip on 3dfx card
• The Mainboard ain’t nuked too. Only replace Bios chip.

Possible causes

• Bios chip has a failure.

Solution

• Replace Bios chip
• Program it with correct Bios ROM (V1.06 i.e.)
• Test the card.

Possible causes

• TEMPLATE

Solution

• TEMPLATE

Possible causes

• TEMPLATE

Solution

• TEMPLATE

Possible causes

• TEMPLATE

Solution

• TEMPLATE