Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Use of QuickSwitches on MAXII and some other PCI Compliant Dev. Kits (Stratix, etc.)

Status
Not open for further replies.

Sylver

Newbie level 2
Joined
Jul 31, 2016
Messages
2
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
36
Can someone explain to me why there are 5V to 3V QuickSwitches on the 50some PCI Bus Signals/Lines on the MAXII Development Board Kit while this CPLD (EPM1270) is equipped with PCI 3.3V Clamp Diodes ???

I found out this is the case with other Altera Dev. Kits too! (at least the Stratix Development Board)
(see the Schematics at the end of this document: https://www.altera.com/content/dam/a...tix_pci_bd.pdf )

This seems redundant to me.
Unless I'm missing something ???

Since the EPM1270 is equipped with PCI 3.3V Clamp Diodes, why not connect the 5V Bus Signals/Lines directly to PCI pins of the EPM1270 ???

Thanks for your lights on this matter because we're about to design a PCB supporting the 5V PCI Bus based on the MAXII Device...
 

AFAIK, PCI 3.3V clamp diodes are for overshoot not for clamping 5v down to 3.3V (which is level shifting) you'll probably burn out the clamp diodes due to the resulting current caused by the mismatched voltage levels.
 
  • Like
Reactions: Sylver

    Sylver

    Points: 2
    Helpful Answer Positive Rating
Thanks for your reply 'ads-ee'.

If I follow you well, the Clamp Diodes present on the PCI Compliant Bank of the MAX II and other similar devices are there to protect only from "overshoots" which seems to be very small surges of voltage over 3.3V (maybe tenths of volts???) ?

If this is the case, this make me think that the MAX II is much more intolerant to current flow (or wattage?) than to the voltage itself. I say this because AFAIK, the diode will surely clamp the voltage to 3.3V as expected, but at the expense of in excess current allowed for the device's PCI pins (which seems to be 20ma in Output mode but could not find in Input mode???).

Also, if "overshoot" protection is so important, why not protecting all pins of the MAX II device (LVTTL, LVCMOS, etc.), not only the PCI ones ?

This being said, I saw some suggest to put small Resistors on these PCI Signals lines in place of QuickSwitches, but I cannot see how this can be done without affecting (dropping) the strength of the PCI Signals voltage in OUTPUT mode?

Regarding the Level Shifters solution (which I think is the same thing than the Quickswithes (i.e. IDTQS3861 or SN74CBTD3384)), I have 1 comment and 1 question:

1) Despite the fact that Altera has published a document explaining how to use these Quickswithes to interface 3.3V devices with a 5V PCI Bus ("Connecting Altera 3.3-V PCI devices to a 5-V PCI Bus", https://www.altera.com/content/dam/altera-www/global/en_US/pdfs/literature/an/an330.pdf), the document states a series of notices on page 3 discouraging the use of these switches for not being fully PCI Specifications compliant...

2) How the use of Level Shifters on every PCI Signals lines will affect or impact the PCI 5V requirement in OUTPUT mode? In other words, will the 3.3V OUTPUT level strength be sufficient to satisfy the INPUT requirements of the 5V PCI Specifications for Vih ?


Hope you 'ads-ee' or any other forum member can react to my comments and/or answer to my questions...

Thanks for your help!

Sylver
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top