Article says Most Flashcarts Damage Consoles. Confirm?

Discuss hardware-related topics, such as development cartridges, CopyNES, PowerPak, EPROMs, or whatever.

Moderator: Moderators

Post Reply
User avatar
TR3KT
Posts: 40
Joined: Sat Oct 01, 2016 2:20 am
Location: Australia, the land of British 50hz and obscure sports games.

Article says Most Flashcarts Damage Consoles. Confirm?

Post by TR3KT »

Hello everybody!
I have stumbled across an from a site called: "DB electronics". They produced this article: The Dangers of 3.3V Flash in Retro Consoles.

So this article makes claims that some Everdrives, if not, most cause major hardware issues and can permanently damage your console, i would like this peer-review and confirmed if this is in fact: True. and the use of such flashcarts can shorten the life of NES and SNES consoles. I would also like to see if the Flashcarts in this list that confirmed to be usable are in fact, fault free. Anyone with knowledge in such issues, feel free to debunk or confirm.
User avatar
FrankenGraphics
Formerly WheelInventor
Posts: 2064
Joined: Thu Apr 14, 2016 2:55 am
Location: Gothenburg, Sweden
Contact:

Re: Article says Most Flashcarts Damage Consoles. Confirm?

Post by FrankenGraphics »

Regarding the diode side of the article:
This is not necessarily to say it is good practice, but clamping diodes are in the world of modular synthesizers sometimes not only used for protection against short/irregular high voltages but also used as a means to modify control voltages and audio alike, so such a pcb would see much more heat dissipation from diodes because of them more ore less continously clamping. But - also in that scene, people are generally more savvy at replacing components, should they fail.

I do think there is some wiggle room - if the heat is within specs, the diodes will likely be fine*.

I think the author was a bit unclear on PCB:s omitted from the list. If he know ones that are good to go according to his judgment, they should be written down. A quick list of PCB:s he knows of but haven't tested would also be handy.

*layman statement. I do not have an EE degree, nor is my autodidact knowledge adequate.
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Article says Most Flashcarts Damage Consoles. Confirm?

Post by lidnariq »

The article is more or less completely accurate.


There's one very specific and narrow exception, which (AFAIK) doesn't actually cover any consoles in that listing. If there are any consoles that are purely NMOS (2600? VIC-20? C64?), then (and only then) would ignoring voltage translation be ok. (last time I said this)

Code: Select all

                   
.................        
        5V       :        ..............
        |        :       :              
        |  nmos  :       :              
    | +-+        :       :     3.3V       
 ---| |          :       :      |       
    | +-+        :       :      _       
        |        :       :      ^       
        |        :       :      |       
        +-----------------------+--...  
        |        :       :      |       
    | +-+        :       :      |       
 ---| |          :       :      |       
    | +-+        :       :      _       
        |        :       :      ^       
        |        :       :      |       
       gnd       :       :     gnd      
console          :       : rom/fpga     
.................         ..............
nMOS pullups aren't very good at finishing things: they can't drive the output all the way up to the voltage supply. The MOSFET ceases to conduct once the difference between the voltage on its input ("gate") and output ("drain") is less than the "threshold voltage". So when the 2A03 is driving a pin, it will pull up strongly to about 4V and then stop. 4V is only 0.7V above the 3.3V supply, or roughly the expected forward voltage of the overvoltage protection diode. So, in this case, it'll conduct only a small amount of current without even needing a current limiter. It'd still be better (if supported by the 3V devices) to run them at 3.6V.

Note that the NES's RAMs are still CMOS, and the caveats in that article still apply to them.
User avatar
tokumaru
Posts: 12427
Joined: Sat Feb 12, 2005 9:43 pm
Location: Rio de Janeiro - Brazil

Re: Article says Most Flashcarts Damage Consoles. Confirm?

Post by tokumaru »

Oh fuck, that sucks. At least my NES and my SNES are safe. I wonder what the verdict is for the PowerPak and for the Harmony.
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Article says Most Flashcarts Damage Consoles. Confirm?

Post by lidnariq »

I strongly suspect the 2600 is purely NMOS, so the harmony cart is probably "ok" for the reason I gave above.

The FPGA in the PowerPak (XC2S30) can apparently be configured to be 5V tolerant or not at fusemap generation time???? (And all other parts on the board are 5V)
User avatar
Drew Sebastino
Formerly Espozo
Posts: 3496
Joined: Mon Sep 15, 2014 4:35 pm
Location: Richmond, Virginia

Re: Article says Most Flashcarts Damage Consoles. Confirm?

Post by Drew Sebastino »

Good to see the money from the high cost of the SD2SNES is being put to good use. :lol: I have to ask, what inherently makes something a specific voltage? I thought with electronics, the voltage something is rated at is the highest it will take without being damaged overtime. Shouldn't this also damage the flashcart? (I think you people already know how knowledgeabe I am about
electronics from my Gameboy Color project...)
calima
Posts: 1745
Joined: Tue Oct 06, 2015 10:16 am

Re: Article says Most Flashcarts Damage Consoles. Confirm?

Post by calima »

If something is rated for X voltage, its components will likely overheat given a higher one, releasing the magic smoke that makes it work. In other words, they were designed to only take that much punishment.

A higher voltage also means it may jump more easily across air gaps, if the design is tight enough. Other failure modes exist, depending on the component.
User avatar
FrankenGraphics
Formerly WheelInventor
Posts: 2064
Joined: Thu Apr 14, 2016 2:55 am
Location: Gothenburg, Sweden
Contact:

Re: Article says Most Flashcarts Damage Consoles. Confirm?

Post by FrankenGraphics »

I have to ask, what inherently makes something a specific voltage?
If you're asking how voltage, current and resistance relates to one other, it's Ohm's law.

This article explains it:
https://www.allaboutcircuits.com/textbo ... ce-relate/

If you're asking what causes a voltage to exist, think of electrons as water and voltage as pressure. The electrons "want" to be evenly distributed. But/so if there's pressure from a reservoir or surge from a drain, it will cause a stream. voltage is a measure of surplus and/or deficiency of electrons. It is measured as a difference between two points.

After reading, you may also want to look up a nifty little basic circuit block called a voltage divider, which is essentially a measured fork in the piping.

Edited many times in attempts to be more clear.
Last edited by FrankenGraphics on Mon Jul 17, 2017 2:36 am, edited 5 times in total.
User avatar
krzysiobal
Posts: 1037
Joined: Sun Jun 12, 2011 12:06 pm
Location: Poland
Contact:

Re: Article says Most Flashcarts Damage Consoles. Confirm?

Post by krzysiobal »

How great my flashcart is free of this issue due to 5V -> 3.3V converters ;)
viewtopic.php?f=9&t=15757
Great Hierophant
Posts: 780
Joined: Tue Nov 23, 2004 9:35 pm

Re: Article says Most Flashcarts Damage Consoles. Confirm?

Post by Great Hierophant »

lidnariq wrote:The article is more or less completely accurate.


There's one very specific and narrow exception, which (AFAIK) doesn't actually cover any consoles in that listing. If there are any consoles that are purely NMOS (2600? VIC-20? C64?), then (and only then) would ignoring voltage translation be ok. (last time I said this)

Code: Select all

                   
.................        
        5V       :        ..............
        |        :       :              
        |  nmos  :       :              
    | +-+        :       :     3.3V       
 ---| |          :       :      |       
    | +-+        :       :      _       
        |        :       :      ^       
        |        :       :      |       
        +-----------------------+--...  
        |        :       :      |       
    | +-+        :       :      |       
 ---| |          :       :      |       
    | +-+        :       :      _       
        |        :       :      ^       
        |        :       :      |       
       gnd       :       :     gnd      
console          :       : rom/fpga     
.................         ..............
nMOS pullups aren't very good at finishing things: they can't drive the output all the way up to the voltage supply. The MOSFET ceases to conduct once the difference between the voltage on its input ("gate") and output ("drain") is less than the "threshold voltage". So when the 2A03 is driving a pin, it will pull up strongly to about 4V and then stop. 4V is only 0.7V above the 3.3V supply, or roughly the expected forward voltage of the overvoltage protection diode. So, in this case, it'll conduct only a small amount of current without even needing a current limiter. It'd still be better (if supported by the 3V devices) to run them at 3.6V.

Note that the NES's RAMs are still CMOS, and the caveats in that article still apply to them.
In his second article and the accompanying Youtube video, Rene specifically used the Genesis S-RAM as an example of an affected component. These S-RAMs should be CMOS like the NES's S-RAM. Curiously, the early Genesis and Mega Drive consoles used the MC68000s and their clones for the most part until Sega removed the High Definition Graphics label from the Model 1s. Then they used the CMOS MC68HC000. Additionally, by the Model 2s they were using the CMOS Z84C000s instead of the NMOS Z084000s. I suppose that means that earlier Genesis systems are safer, at least as to some components, than later ones.

According to wikipedia, the DRAM in the C64 is CMOS but the rest of the chips seem to be NMOS, at least for the earlier units.
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Article says Most Flashcarts Damage Consoles. Confirm?

Post by lidnariq »

calima wrote:A higher voltage also means it may jump more easily across air gaps, if the design is tight enough.
I'm under the impression that lower-voltage-rated ICs are using thinner layers of dielectric (quartz, in this case), and so using too high of a voltage could cause a dielectric breakdown on the MOSFET gates.
Great Hierophant wrote:According to wikipedia, the DRAM in the C64 is CMOS but the rest of the chips seem to be NMOS, at least for the earlier units.
I'm pretty certain the ones in this picture are NMOS DRAMs ... at least the data sheet I can find for the HM4864 only mentions "Hitachi's double-poly N-channel silicon gate process" and, at the time, advertising things as being explicitly CMOS seemed to be de rigeur.
User avatar
mikejmoffitt
Posts: 1353
Joined: Sun May 27, 2012 8:43 pm

Re: Article says Most Flashcarts Damage Consoles. Confirm?

Post by mikejmoffitt »

Sorry, I'm just a little confused. Isn't Great Hierophant's lidnariq's drawing on the left one of a CMOS output, sinking and sourcing?
Last edited by mikejmoffitt on Tue Jul 18, 2017 1:50 am, edited 2 times in total.
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Article says Most Flashcarts Damage Consoles. Confirm?

Post by lidnariq »

You mean the one quoted from me?
The following ascii art would mean nMOS to me:

Code: Select all

       |         |          |
   | +-+     | +-+      | +-+
 --| |     --| |<-    --| |  
   | +-+     | +-+      | +>+
       |         |          |
I'd personally use one of these to mean pMOS:

Code: Select all

       |        |           |
   | +-+    | +-+       | +<+
 -o| |    --| |->     --| |  
   | +-+    | +-+       | +-+
       |        |           |
I don't have a good way of distinguishing between enhancement and depletion-mode MOSFETs in ascii art. I think I'm ok with that.
Post Reply