13-04-2011, 04:53 PM
ESR meter construction, testing, setting up and troubleshooting notes.
In my first post, the picture of the meter that I attached had no meter movement in it is they’re out of stock at ESR Electronics just now. Hence, to give an idea what the finished meter looks like, I’ve attached a pic of my original one, which had the zero adjustment pot on the front panel, but which is now a preset pot on the PCB.
These additional notes, based on help that I've given to several constructors who have had teething troubles building the meter, might help less experienced constructors to build and set up the project, and if need be, to troubleshoot and get it working. As with all projects it’s important to check that components are the correct value, particularly those which have obscure markings. Even resistor markings aren’t so straightforward these days - often using the 5-band marking code, as at Maplin for example. Also, ensure that diodes are connected the right way round.
The 2N222A transistor comes in two case styles, the more common of which is the metal ‘TO18’ style case. It also comes in the plastic TO92 case. On the ‘x-ray’ picture of the updated PCB, I’ve shown the pin-outs of each case style.
A word of warning though – there are less common TO92 cased 2N222A transistors in which the emitter and collector are reversed. It’s therefore wise to use a metal cased one, or to check the connections on a transistor tester to be sure.
Setting up and troubleshooting:
Initial tests and setting up:
On completion of the project, set the meter adjust pre-set pot on the PCB to midway position, and before inserting the IC, switch the meter on, and check that there is 5V present at pin 14 of the IC socket. (Black test prod to pin 7 of the socket – red to pin 14). If all’s well, insert the IC, clip the test prod leads to each other, and the meter needle should deflect to the right. If so, adjust the preset pot on the PCB until you get full scale deflection on the meter. It should not need adjusting again.
Calibrating the dial:
Having zeroed the meter with the pre-set pot so that with the test prods shorted it shows full-scale defection - zero Ohms, you can begin calibrating the dial using a few low value resistors as follows:
Two 1 Ohm resistors in parallel = 0.5 Ohm, one on its own = 1 Ohm, two in series two Ohms, then more 1 Ohm resistors up to 10 Ohms, and so on. On the 50uA meter that I used 1 Ohm was about 80% of full scale, 5 Ohms mid-scale, 15 Ohms about 30%. Really, any cap approaching 10 Ohms is highly suspect, so it’s the low Ohms end we’re interested in, at the RH end of the scale.
What if it doesn’t work?
Carefully check all the components, especially the correct orientation of the diodes, Q1, the voltage regulator, the IC and C7 (the only polarity sensitive capacitor). Using a magnifying glass, check for any solder bridges - for example between the pins of the IC socket. Check that you have 5V+ at pin 14 of IC1. If not, is there 9V at the input of the voltage regulator? If so, suspect the regulator. Check that you've used the right IC - 74HC14N. (Not 74HCT14N). It helps if a scope is to hand to check the waveform at various points, and to gain an understanding of the various elements of the circuit.
Getting the meter to read Full Scale Deflection:
The original circuit specified a 10k resistor for the 50 uA meter shunt, (R17) though the designer refers to having to reduce that to 4k7. I found that I needed to reduce it still further to 3k3. Make sure you’ve used a 3k3 resistor for R17 - not a 10k resistor for R17 as was in the original design.
Though the elements of the circuit aren't complex, it does help in troubleshooting if a scope is to hand, but it's worth stressing that when testing and setting up the device, unless the test leads are shorted out, there will be no deflection on the meter and 0.0mvpp at C2 into the AC amplifier (Q1 etc). I’ve attached some pics to show the waveform at various parts of the circuit in a working model (with the test leads shorted), and have shown the scope switch settings.
The pics a are:
1) At pin2 of IC1 - the output of the oscillator/waveform generator, (which simply consists of one resistor, one cap, and one gate of the hex Schmidt trigger IC).
2) At the base of Q1
3) At the collector of Q1
If the meter doesn't work, it's unlikely to be the transistor or IC that are at fault as they're not especially delicate. But as the ICs are just a few pence each, it's easy enough to swap one to see if the original is indeed dud.
From my post-bag over the least few months, it's clear that lots of these little gizmos have been built and are in use, and though it's not a complex project to build or set up, quite a few constructors have had difficulties. Invariably it's been due to shorted tracks, dry joints, polarity of diodes wrong, wrong value components (by a factor of 10 or 100), IC not inserted correctly, or the meter movement connected with reversed polarity. It’s wise to test all components such as resistors and caps before insertion to check that their values are correct.
If anyone does build this project, and after checking all of these points, still has difficulty getting it to work, I’ll be happy to help and advise.
Hope that’s of interest.
David
In my first post, the picture of the meter that I attached had no meter movement in it is they’re out of stock at ESR Electronics just now. Hence, to give an idea what the finished meter looks like, I’ve attached a pic of my original one, which had the zero adjustment pot on the front panel, but which is now a preset pot on the PCB.
These additional notes, based on help that I've given to several constructors who have had teething troubles building the meter, might help less experienced constructors to build and set up the project, and if need be, to troubleshoot and get it working. As with all projects it’s important to check that components are the correct value, particularly those which have obscure markings. Even resistor markings aren’t so straightforward these days - often using the 5-band marking code, as at Maplin for example. Also, ensure that diodes are connected the right way round.
The 2N222A transistor comes in two case styles, the more common of which is the metal ‘TO18’ style case. It also comes in the plastic TO92 case. On the ‘x-ray’ picture of the updated PCB, I’ve shown the pin-outs of each case style.
A word of warning though – there are less common TO92 cased 2N222A transistors in which the emitter and collector are reversed. It’s therefore wise to use a metal cased one, or to check the connections on a transistor tester to be sure.
Setting up and troubleshooting:
Initial tests and setting up:
On completion of the project, set the meter adjust pre-set pot on the PCB to midway position, and before inserting the IC, switch the meter on, and check that there is 5V present at pin 14 of the IC socket. (Black test prod to pin 7 of the socket – red to pin 14). If all’s well, insert the IC, clip the test prod leads to each other, and the meter needle should deflect to the right. If so, adjust the preset pot on the PCB until you get full scale deflection on the meter. It should not need adjusting again.
Calibrating the dial:
Having zeroed the meter with the pre-set pot so that with the test prods shorted it shows full-scale defection - zero Ohms, you can begin calibrating the dial using a few low value resistors as follows:
Two 1 Ohm resistors in parallel = 0.5 Ohm, one on its own = 1 Ohm, two in series two Ohms, then more 1 Ohm resistors up to 10 Ohms, and so on. On the 50uA meter that I used 1 Ohm was about 80% of full scale, 5 Ohms mid-scale, 15 Ohms about 30%. Really, any cap approaching 10 Ohms is highly suspect, so it’s the low Ohms end we’re interested in, at the RH end of the scale.
What if it doesn’t work?
Carefully check all the components, especially the correct orientation of the diodes, Q1, the voltage regulator, the IC and C7 (the only polarity sensitive capacitor). Using a magnifying glass, check for any solder bridges - for example between the pins of the IC socket. Check that you have 5V+ at pin 14 of IC1. If not, is there 9V at the input of the voltage regulator? If so, suspect the regulator. Check that you've used the right IC - 74HC14N. (Not 74HCT14N). It helps if a scope is to hand to check the waveform at various points, and to gain an understanding of the various elements of the circuit.
Getting the meter to read Full Scale Deflection:
The original circuit specified a 10k resistor for the 50 uA meter shunt, (R17) though the designer refers to having to reduce that to 4k7. I found that I needed to reduce it still further to 3k3. Make sure you’ve used a 3k3 resistor for R17 - not a 10k resistor for R17 as was in the original design.
Though the elements of the circuit aren't complex, it does help in troubleshooting if a scope is to hand, but it's worth stressing that when testing and setting up the device, unless the test leads are shorted out, there will be no deflection on the meter and 0.0mvpp at C2 into the AC amplifier (Q1 etc). I’ve attached some pics to show the waveform at various parts of the circuit in a working model (with the test leads shorted), and have shown the scope switch settings.
The pics a are:
1) At pin2 of IC1 - the output of the oscillator/waveform generator, (which simply consists of one resistor, one cap, and one gate of the hex Schmidt trigger IC).
2) At the base of Q1
3) At the collector of Q1
If the meter doesn't work, it's unlikely to be the transistor or IC that are at fault as they're not especially delicate. But as the ICs are just a few pence each, it's easy enough to swap one to see if the original is indeed dud.
From my post-bag over the least few months, it's clear that lots of these little gizmos have been built and are in use, and though it's not a complex project to build or set up, quite a few constructors have had difficulties. Invariably it's been due to shorted tracks, dry joints, polarity of diodes wrong, wrong value components (by a factor of 10 or 100), IC not inserted correctly, or the meter movement connected with reversed polarity. It’s wise to test all components such as resistors and caps before insertion to check that their values are correct.
If anyone does build this project, and after checking all of these points, still has difficulty getting it to work, I’ll be happy to help and advise.
Hope that’s of interest.
David
