Residential write-up - Kevin Moore
Residential Write-up - Mark Hankins and Ben Surtees
Whilst taking part in the scheme, participants are expected to put in quite a lot of hours. Most of this is done as part of a team, it is vital therefor that the team should get to know each other and have a chance to talk through ideas together. To help to do this, as part of the scheme, the teams are expected to attend a residential weekend at a university. This year we went to the residential at Newcastle University, this gave us a chance to use any of the university facilities, and because it was a 3 day residential we stayed in a nearby hotel. This was not only educational, but was also a great weekend, we all had a good laugh and everyone got to know their engineers and team members much better. After the weekend we all had a much better understanding of the project because we had spent 3 days solid talking about nothing else (Well a few other things...but never mind that!)
Contrary to popular belief, we did actually get quite a lot of work done at the residential, we all wrote reports about what we had done while we were there, so most of what we have written is very accurate. We got most of our prototyping done when we were there and we bombarded Andrew Lamb (Our designated engineer) with questions. It must have been quite infuriating for him, but that is what he was there for.
Residential write-up - Kevin Moore
"I have connected up my circuit on breadboard at the university. The resistor and the capacitor values are so that the pulses are reasonably slow so that they can be viewed on the oscilloscope easily. At the point A on the circuit diagram a ramp voltage is produced. This is made up of 16 steps. I was hoping for a 'ramp' effect, I removed an extra op-amp from the circuit and this magically appeared as I expected."
When the ramp voltage is higher than the input voltage at the inverting input, the output of the op-amp will go high. This is used as a signal to the processor which, when the signal is received, the output from the binary counter will be recorded and stored in memory. This is then displayed on the LCD.
When I tested this circuit my results were reasonably encouraging. I managed to show two traces on the oscilloscope screen of the ramp voltage and the output voltage. The output voltage, however, followed exactly the same pattern, regardless of the input voltage to the op-amp as soon as the voltage exceeded a point around 4V. This would be a huge disadvantage to the circuit as the Tx+ line runs off a theoretical voltage of 5V.
This is the stage that I have reached by the end of Wednesday, the first day of the residential.
I asked the advice of our engineer, Andrew Lamb, on my problem this morning (Thursday). He suggested that the problem lies with the type of op-amp that I am using. As a result of the internal resistance of the op-amp, there is a voltage drop of around 1V. After looking at the specification points of the op-amps that I am using, I have found that the output voltage is quoted to range from ground to the positive supply voltage – 1.5V. This explains the problem.
I suggested that one way to solve the problem would be to use a different voltage so that even after a voltage drop, the voltage would be above the required 5V. In practice however, this would not be sensible, as another battery would be required.
The best solution would probably be to use a different op-amp which is more suitable to the situation. Op-amps are available that will produce a voltage almost equal to the supply. Unfortunately, the first op-amps that I was going to use will only work, in theory, to within 1.5V of the positive power supply. He said that he would try to bring some new op-amps tomorrow. Until then I will try to understand more about the circuit that I am using."
Kevin Moore - Before, During and After the residential.
Residential
Write-up - Mark Hankins and Ben Surtees
What we planned to do:
By the end of the week we hoped to have both the battery monitor circuit and the surge protection circuit working on breadboarding order to be able to carry out tests on both.
What we did:
Over the first three days of the week, we managed to breadboard the surge protection circuit. We carried out initial testing on this and it appeared to work. We still needed to do further testing to see if it would work in the required situation, but we did not have the facilities or the equipment to test everything because this would mean that we would have to put a huge load across the circuit and see if it reacted as it was supposed to. i.e. the resistors would heat up, their resistance would increase and they would heat up more, this produces a snowball effect, which stops the rest of the circuit from being wrecked.
We also breadboarded the circuit for the battery charger/monitor circuit. We needed the Thyristor (2N5060) to be able to complete the circuit. We were awaiting the part being delivered on Thursday (This part was not available at the university so it had to be picked up from Tynetec and delivered to us.)
Thursday 6th January
We need to test the surge circuit and the battery monitoring circuit to check that both of them work. Once we had conducted a little initial testing, we went over to the university library and had a look around to see if we could find any circuits which already existed. We could then decide if these would be more efficient than the one we had developed. If we could find one then we could hopefully use it, or at least adapt it to enhance our own circuit. Unfortunately, because the library is really there for the use of the university students, all of the publications were too advanced for us and we did not have enough background knowledge on the subject to be able to understand what most of the lierature was talking about. This meant that we left the library with very little additional information compared to when we went in. It was did give us an insight into just how much information was available here if we ever needed it.
While we were initially testing the battery circuit, we thought that it was not working, we decompiled the circuit and then rebuilt it, only to find that it was working exactly the same. We went over the circuit diagram to check that it should work properly in theory, and we found that we had mad a mistake, this would mean that the circuit was working in reverse, this giving us a full battery indicator instead of telling us when the batter was nearly finished. We corrected the error and the circuit began to work as we had planned.
We decided to try to test the surge circuit by simulating a massive voltage spike. We did this by plugging it into a power supply and then increasing the power. We then realized that testing like this was not really appropriate because in a real voltage spike the voltage level would dramatically increase over such a small space in time that we could not really accurately simulate this by hand. We did not have access to the sort of voltages which would be applied in a real situation (e.g. a lightning strike!)
In the battery monitoring circuit, we will be alerted to the state of the batter when the voltage is higher than 6.5v or higher, or when it is at 5.7 volts or lower.
Home
Acknowledgements
Design Brief
Technical Details
Residential
Pictures