QUOTE (pagercam @ Jun 25 2005, 12:22 AM)
I think its a bit inappropriate to call a device bad engineering when an untrained technician dismantles a product and it doesn't work in another configuration. Benq knows that thier monitors will be used as monitors with all the sheilding in place and has no expectation that it will be used in any other way. The sheilding provies the continuous ground and that is one way to do the engineering, and as an engineer I have no problem with that way to provide ground. As many of the people at LL don't have an engineering or electrical backround I'll explain a few things that may help people understand the issues involved.
There are two parts to DC power, the power line and the ground, it is the most important thing, especially in digital electronics, that there is one and only one ground so that all circuits (chips or boards) have their signals referenced to the same ground potential. There is a joke, engineering jokes aren't funny to non-engineers, that says that digital engineers have only heard rumours of something between zero and one, i.e. volatges are either high or low (0 or 1), but what this really means is that if the volatge is above one level it is considered high (1) and below another volatge it is considered a zero. Different types of chips have different thresholds, but in all cases there are volatges that are neither high nor low and the objective of every digital engineer is to make sure that there is enough enough power to make sure that the circuits can be driven to the high voltage, poor design or overloading can drop a volatge into the not high not low volatage area. These voltage thresholds are in reference to ground so if we are using 5 volt logic where high is anything over 3 volts and low is anything below 2 volts, 2.5 volts would be invalid and nothing works, but 3 volts on a circuit connected to one ground may only be 2.5 volts connected to a circuit using a different ground.
In good design you try and limit the use of voltages to specific areas to minimize the amount of noise being generated by one circuit interfereing with another so voltages are often seperated between boards but those two boards can not communicate at all unless they are at the same ground potential. There are multiple ways to distribute ground but it is risking to provide a ground to one board and then pass that ground to another board and another, as there is no guarentee that the ground coming into the system will be at the same level after it has passed from one board to another, so optimal design says that all boards should be connected to the same ground but this can make the wiring more difficult. To limit the extra wiring for ground a chasis ground is provided so that the RF sheild that is required by FCC regualtions and is at ground protential also serves as an electrical ground its a big surface and makes a good ground. This however makes the proper working of the circuits depenant onbeing connected to the sheilding to work well. Runing wires directly from the power supply to all the boards is another way to do it or carefull daisy chaining between boards can work if it is done carefully, they all work its a matter of preference which any engineer uses on an specific project.
So I can't fault the design, its perfectly valid. We don't have the design details of each of these monitors, so we don't know which monitors can be used one way or another. If a board was designed to daisy chain grounds between boards, connecting all the grounds could cause problems, and some people at LL have report monitor problems until they removed the ground wires. Others have reported problems without ground wires and this is just a refelection on different design styles not bad engineering.
Having multiple grounds is as bad as not having grounds because it can generate what are known as ground loops in which having two connections to ground makes the ground unstable as it tries to balance the potentail between the two sources.
So you have to understand the design to know how to modify it and taking it out of its casing is modifying it.
I understand that you were incrediably frustrated by having an monitor that appeared dead but it wasn't Benq's fault that you changed their design. By the way, you're use of connecting the same board on multiple points coould lead to the ground loop situation so I suggest that connect the boards with grounding wires at one and only one point each.
Pager,
First off, thank you for the insightful and thoughtful reply. I am sure that it will be useful to people in the future who have limited experience with electronics.
I thought a bit before responding to your post. I wasn't going to, but then it left me feeling a little deflated and I thought I should set the record straight. This is a forum for people to ask for help and share their experiences and opinions regarding all things building projector-related. This is exactly what I did. If you were offended by my opinion, you could have personal messaged me with your concerns, and perhaps I could have had the post deleted.
Instead, you refer to me in your post as "an untrained technician". I don't want to turn this into a "show me your credentials" thread, but I also have a degree in Electrical Engineering from Cal, Berkeley; I know whereof I speak.
In my opinion, using the chasis to carry dc common is an invitation to induce noise. The chassis acts like an antenna. There are traces on the board for the specific purpose of carrying ground. Granted, my opinion was colored by the nightmare I experienced where my BenQ worked flawlessly without grounds connected, and then a couple of hours later, refused to even power on. Indeed, many here have been thrilled with their BenQ's.
I felt, however, rather belittled after reading your post and thought I should communicate that to you publicly. If this was not your intention, then I apologize.
Try toubleshooting with just the PSU, LCD controller, and control panel to see if you can get it to power up on it's own.
:angry: 
) - This is the crappiest example of engineering I have seen to date. The darn thing worked flawlessly without the grounds and then decided to not turn on altogether. Even if the unit appears to work without these connection, I would attach them as a matter of course with this LCD.