Why do some signal rerouting situations require cutting the existing connection?

In summary, the concept of a component being "drop in" or not is important when replacing or modifying signals on a PCB. If it is not a drop in component, the traces and routing may need to be changed in order to provide new or modified signals to the new component. In some cases, it is easier to cut a signal off from an existing path rather than removing or resoldering a component. This is often done when the new signal source would conflict with the old signal source. However, the only way to completely remove a signal is to cut its electrical connection.
  • #1
Xzyx987X
6
0
This is probably a complete noob question, but due to my general lack of experience in this area I wasn't able to find the answer on my own. When you want to reroute a signal on the PCB to an external location, or replace a signal with one of your own, what situations do you need to cut the signal off from an existing path? For instance, in this example, a BIOS chip is successfully overridden by soldering a new chip over the existing chip with no cuts made. In many console modchips (although I don't know what their precise functionality is) many signals are rerouted and modified just by soldering to the surface of the PCB. But in http://devcast.dcemulation.com/mods/overclock/overclock.php example says you need to cut the pin to replace the clock signal. So why exactly is it that in some cases the signal can be overridden by soldering over a connection, and why is it in others you have to cut the existing connection before you can add a new one? Also, is it possible to pull off the mod in the last example without making any irreversible cuts?
 
Engineering news on Phys.org
  • #2
You must learn the concept of a component being "drop in" or not. If it is drop in, it means that when you replace one component that is drop in compatible with it, it uses the exact same signals and often the exact same footprint as the original component, although its functionality may be different. If it is not a drop in, then the board's traces and routing must be omitted, added, or connected in a way different than the original layout provides so that you can provide new or modified signals to the new component.

Think of your PCB as a breadboard in this situation, and if you pulled one component out and put in a new one, is the schematic and/or layout going to remain the same or does it need to change? If everything remains the same, then its like your example of the BIOS chip.

The reason they are cutting that pin in the other example is because you still want to keep that chip soldered onto the board, and to have to resolder or remove it would be a pain, so its easier to just cut that pin since that is the only signal they need to change. This now gives them access to that signal to put the new clock source on without having the original clock source trying to drive the signal at the same time as your new clock.

What would happen if you just soldered the new clock source to that pin without cutting it? You would have two clock sources trying to drive the same signal.
 
  • #3
TheAnalogKid83 said:
You must learn the concept of a component being "drop in" or not. If it is drop in, it means that when you replace one component that is drop in compatible with it, it uses the exact same signals and often the exact same footprint as the original component, although its functionality may be different. If it is not a drop in, then the board's traces and routing must be omitted, added, or connected in a way different than the original layout provides so that you can provide new or modified signals to the new component.

Think of your PCB as a breadboard in this situation, and if you pulled one component out and put in a new one, is the schematic and/or layout going to remain the same or does it need to change? If everything remains the same, then its like your example of the BIOS chip.

Hmm... But then, why is it that the chip soldered on the top ends up being the one the data is read from and not the one underneath that you soldered over?

TheAnalogKid83 said:
The reason they are cutting that pin in the other example is because you still want to keep that chip soldered onto the board, and to have to resolder or remove it would be a pain, so its easier to just cut that pin since that is the only signal they need to change. This now gives them access to that signal to put the new clock source on without having the original clock source trying to drive the signal at the same time as your new clock.

What would happen if you just soldered the new clock source to that pin without cutting it? You would have two clock sources trying to drive the same signal.

Well, that much makes sense. But is there any way you can think of to get rid of the other clock signal without cutting the connection?
 
  • #4
Xzyx987X said:
Hmm... But then, why is it that the chip soldered on the top ends up being the one the data is read from and not the one underneath that you soldered over?

It looks like he says that some of the memory control lines need to be rewired and to email him about the specifics. Also, one of his photos is labeled "chip selects" and shows the two chips stacked with some wires soldered to only one of the chips, which I would take implies that the two flashes have different chip selects. This would mean that only one chip is accessed on the memory bus at a time, and the unused flash chip would be in high-z state while the other is accessed.


Xzyx987X said:
Well, that much makes sense. But is there any way you can think of to get rid of the other clock signal without cutting the connection?

It is an electrical connection. You need to remove the electrical connection, you can't get around the physics of shorting two signals together electrically. The easiest way would be to cut the pin as in the example, thus the chip would no longer be electrically connected to the rest of the node. The other way would be to unsolder the pin and bend it up or just remove the chip completely if it isn't necessary to the functionality of the circuit anymore. The only other way would be to disable the chip and put the clock driving pin in a state that would not affect the rest of the circuit, although that is a little far-fetched.
 
  • #5
Ok, so what about modchips, which override signals by only soldering to various points on the PCB. How do those work without having to cut connections?

Edit: I see what you mean with the BIOS chip. I didn't look at that one closely, just pulled it out because I happened to remember it as it also was a Dreamcast project..
 
Last edited:
  • #6
Xzyx987X said:
Ok, so what about modchips, which override signals by only soldering to various points on the PCB. How do those work without having to cut connections?

well you have to look at functionality of the pins of the chip. maybe it is input pin, maybe the pin is not used.
 
  • #7
I'm not sure, but http://www.dms3.com/dms4ezi/downloads/Installation_Manual/DMS4_EZI_Installation_v1.4.pdf modchip appears to be overriding the BIOS with nothing but connectors on the chip's pins. How might that work in theory? I suppose this doesn't have much to do with the original question anymore, but I'm still kind of curious. I assume it's by redirecting the signals going into the chip, but how might that theoretically work without cutting the power to the chip or cutting it's intput/output connections?
 
Last edited by a moderator:
  • #8
Xzyx987X said:
I'm not sure, but http://www.dms3.com/dms4ezi/downloads/Installation_Manual/DMS4_EZI_Installation_v1.4.pdf modchip appears to be overriding the BIOS with nothing but connectors on the chip's pins. How might that work in theory? I suppose this doesn't have much to do with the original question anymore, but I'm still kind of curious. I assume it's by redirecting the signals going into the chip, but how might that theoretically work without cutting the power to the chip or cutting it's intput/output connections?

They are not simply putting one chip on top of another, they have an adapter board and connector between the two chips, which means that they could be rerouting on that adapter board. You really need a schematic to know what they're doing, or atleast see the connector device up close and datasheets for the parts and inspect how it connects up close.. to say anything more would be speculation, but its definitely not some magic or unknown technology. Does the connector board with the mod chip have a pull-down/pull-up on one of the pins for the original chip but not on the new one so that they have electrically different connections? This could be on an enable pin for example. Do the chips communication on a serial interface like I2C or SPI to where they could be configured for different addresses on the same bus? There could be a million different simple explanations. You can see that a handful of signals are brought off of these mod connector boards on the colored wires, so you know there's more going on than just the chip "overriding" the other one just by sharing pins. Stick with schematics and circuit knowledge, and you can figure it out easily. Go into the datasheet of the processor or BIOS chip and read the functionality of every signal and I'm sure you could come up with a way to disable it while placing a new one on top of it with the right functionality.
 
Last edited by a moderator:
  • #9
Like I said, I'm still a noob at this kind of stuff, but I basically get the picture. I just wanted to make sure nothing was going on in my examples that defied my current knowledge of electronics. I didn't really think overriding a chip would ever be as easy as putting one chip on top of another, but hardware reverse engineering isn't a topic you can find a lot of information about easily, so I just wanted to be sure.
 
  • #10
Well the best way to know how to reverse engineer is to first know how to engineer in my opinion. I think if you understand how the board works and how the chip works, it makes much more sense how the design can be modified to get a certain result. If you wanted to do some interesting stuff to understand the impact of doing certain things to a circuit, you should play around with schematics and make your own (doesn't have to be your own design, but just connecting all of the components really cements in the ideas of what can happen what can't happen) and maybe try laying out a simple board or running simulations. . a good place to start is here:

www.expresspcb.com

they have free software for you to make your own schematic and then lay it out. I think if you did a couple boards, you'd really get a good idea of how someone can mod a board.
 
  • #11
TheAnalogKid83 said:
Well the best way to know how to reverse engineer is to first know how to engineer in my opinion.
Heh, that's very true. Nearly everything I know about how to hack and reverse engineer software comes from my experiences programming. The problem with applying that methodology to hardware however, is that it can get a bit expensive.

TheAnalogKid83 said:
www.expresspcb.com

they have free software for you to make your own schematic and then lay it out. I think if you did a couple boards, you'd really get a good idea of how someone can mod a board.
Wow, those are some remarkably simple tools. Sometimes people in the technology biz have a knack for making things seem much more complicated than they actually are... I think tools that did simulations would be more useful to me for the moment, but if I ever want to do a PCB layout for manufacturing, I could do worse.
 
  • #12
Xzyx987X said:
Wow, those are some remarkably simple tools. Sometimes people in the technology biz have a knack for making things seem much more complicated than they actually are... I think tools that did simulations would be more useful to me for the moment, but if I ever want to do a PCB layout for manufacturing, I could do worse.

Well those tools were designed for very simple projects or for hobbyists. No commercial product would be designed with those tools or manufactured with that place unless the business was out of someone's basement. Don't get the wrong idea that PCB layout and circuit design are very simple, you can see a lot of problems even on a bread board that make your circuit not work like you think it should.

Some boards are very simple, but others are very complex, like the one's you've been linking to. There's a lot of technology and considerations into the design of these boards and circuits, and I'm surprised some of those mods you linked actually worked because of signal integrity issues. They certainly are going to ruin some of the original design features like EMC. The general ideas on how a PCB works are very easy, but few electronics books or courses teach anything about the actual boards. If you get into it at a bit more depth, you will find all of your physics and electronics knowledge to pay off when it tells you the RIGHT way to lay out a board to ensure that it works reliably and meets stringent guidelines.
 

FAQ: Why do some signal rerouting situations require cutting the existing connection?

What is signal rerouting behavior?

Signal rerouting behavior is a phenomenon observed in various species, including humans, where an individual changes their behavior in response to a change in an expected signal. This can occur when a previously reliable signal is no longer present or when a new signal is introduced.

How do organisms exhibit signal rerouting behavior?

Organisms exhibit signal rerouting behavior through changes in their sensory systems, neural pathways, and behavioral responses. For example, an animal may use a different sense or route to find food if their usual method is no longer available.

Why do organisms exhibit signal rerouting behavior?

Organisms exhibit signal rerouting behavior as an adaptive response to changes in their environment. It allows them to continue to meet their needs and survive in new or changing conditions.

Can signal rerouting behavior be learned or is it innate?

Both innate and learned factors can contribute to signal rerouting behavior. Some responses may be hardwired in an organism's genetics, while others may be learned through experience and trial-and-error.

How can studying signal rerouting behavior benefit us?

Studying signal rerouting behavior can provide insight into how organisms adapt and survive in changing environments. This knowledge can be applied to various fields, such as conservation, animal behavior, and even human psychology and decision-making processes.

Similar threads

Back
Top