Help me troubleshoot this simple design

[whiz115]

I am happy that it is working as that was a lot of work to debug, wasn't it?

i don't care much... besides i learned enough from that!

what i did is to replace R17 with a variable resistor and i'm thinking to leave it there because in case i'm going to use batteries to power the esr-meter i will need to regulate again the output.

you said something about loading down the detector...what do you mean? do i stress the instrument with what i did?

also do you think it can operate with a 9V battery without any problems?

 

[RadioRon]

i don't care much... besides i learned enough from that!

what i did is to replace R17 with a variable resistor and i'm thinking to leave it there because in case i'm going to use batteries to power the esr-meter i will need to regulate again the output.

you said something about loading down the detector...what do you mean? do i stress the instrument with what i did?

also do you think it can operate with a 9V battery without any problems?

Like I said before, be careful that you hook up your variable resistor the right way. There is one way if you are using a VOLT meter at the output and another way if you are using a MICROAMP meter at the output.

Loading does not imply a dangerous condition necessarily. When we talk about a circuit that delivers something to the output, we imagine the Thevenin equivalent circuit for the output. In such an equivalent circuit we know that there is typically a voltage source and a series resistance which is the output resistance of the circuit. Then we we attach a meter or some sort of resistor to the output we call this the load and if we make this resistance lower, we are increasing the load by drawing more current. When we draw more current, the voltage drop across the source resistance increases, so the amount of voltage we can have across our load tends to decrease. This is the effect of loading the output. The degree of it, and whether it is a problem or not, depends on how much larger the load resistance is compared to the source resistance. In simple circuits, we typically want the source resistance to be low and the load resistance to be high so that when we draw more current by gradually lowering the load resistance, the output voltage drops very little.

In our circuit here, the source resistance is not exactly known because it is dependent on how much the rectifiers D1-4 are into conduction, but we can guess that it might be on the order of a few hundred ohms or less when the output voltage is above about 0.2 volts. If, say for example, our source resistance is 400 ohms, then if you put a 10K ohm resistor across C6, you can calculate how much output voltage will drop using Ohm's law. That 10K ohm resistor will have reduced your maximum output voltage by 3.8%.

So, loading down the detector is simply a way of throwing away desireable voltage by putting too low a resistance at the output. It doesn't harm anything unless the thermal, max current or max voltage ratings of parts are exceeded which would not happen in this case.

The circuit is designed to operate at 5 volts. If you attach a 9V battery, the bias applied to Q1 will shift somewhat and the output of Q1 collector will saturate at 0 volts, so this stage would stop working. You can adjust for this by reducing the value of R10 unti the steady state DC voltage (with no capacitor attached to the test points) at Q1 collector is about 4 volts. The other thing that will change is that the threshold at which you get a reading will drop. The original circuit was designed so that you got no reading until the capacitor ESR is about 75 ohms or so, but if you increase the supply voltage, the threshold will drop to about 40 ohms. This might be OK, but it is a change that you must consider as it shifts the scale of the meter.

You can avoid these troubles by adding a 5V regulator between the 9V battery and this circuit. This circuit is drawing about 25 mA or so, and this will increase up to 45 mA or so, which will drain your 9V battery in about 10 hours or thereabouts. It would be better to use 4 alkaline AA cells in series (=6 volts) which needs no regulator.

 

[whiz115]

nigel and the other guys....

today i learned something that may be interesting about ESR meters... it is said that the best way is to use sine wave to test capacitors and not square wave as most esr meters use.

the reason is something that has to do with harmonics which are alot on square waves...

do you think such thing is true?

 

[Nigel Goodwin]

nigel and the other guys....

today i learned something that may be interesting about ESR meters... it is said that the best way is to use sine wave to test capacitors and not square wave as most esr meters use.

the reason is something that has to do with harmonics which are alot on square waves...

do you think such thing is true?

Yes it is, which is why most ESR meters (including all the ones I've seen or used) use sinewaves - I don't know where you've got the impression that they use squarewaves?.

 

[whiz115]

I don't know where you've got the impression that they use squarewaves?.

what about the design on this thread? i think it uses square wave...
can you check the first page?

 

[Nigel Goodwin]

what about the design on this thread? i think it uses square wave...
can you check the first page?

If you read the article it has a low-pass filter after the oscillator to make it more sine-like - but it's a pretty useless design anyway.

 

[whiz115]

can you explain to me why exactly we need perfect sine wave?
my friend couldn't explain to me the reason because he also heard it
from somebody else...

thanks nigel..

 

[Nigel Goodwin]

can you explain to me why exactly we need perfect sine wave?
my friend couldn't explain to me the reason because he also heard it
from somebody else...

You're supposed to be measuring at 100KHz, if you use a squarewave you're measuring at 100KHz plus all the harmonics - the harmonics may be handled completely differently so you can't get an accurate reading.

 

[whiz115]

You're supposed to be measuring at 100KHz, if you use a sinewave you're measuring at 100KHz plus all the harmonics - the harmonics may be handled completely differently so you can't get an accurate reading.

i think you mean squarewave and not sinewave... right?!

 

[ericgibbs]

i think you mean squarewave and not sinewave... right?!

hi,
I'm sure its a typo, should be square.

For pure sinewave use a Wien osc.

Look at this pdf, a member has made this and it works.

 

[Nigel Goodwin]

i think you mean squarewave and not sinewave... right?!

Oops!