Problems building first circuit - capacitor ESR meter

[Rleo6965]

@Jaynuts

I'm just wondering of D2 connection in your provided diagram. It seems dc voltage after D1 will pass directly on forward bias D2 to ground. What would be left voltage to be measured by your uA meter?

 

[(*steve*)]

I believe the idea is that the forward voltage drop of the diode across the meter is there to prevent excessive current from flowing through the meter.

I would not be concerned about the use of a normal rectifier diode here, but D1 should be a fast diode. at almost 50 kHz, a normal rectifier diode isn't going to be doing much rectifying.

 

[Jaynuts]

I believe the idea is that the forward voltage drop of the diode across the meter is there to prevent excessive current from flowing through the meter.

I would not be concerned about the use of a normal rectifier diode here, but D1 should be a fast diode. at almost 50 kHz, a normal rectifier diode isn't going to be doing much rectifying.

Yes my coursework also talks about defining errors, blind alleys and things you can improve on so I will definitely include that in the report.

As for the circuit can someone explain why is it running in astable mode?

Thanks

 

[Rleo6965]

According to Wikipedia about forward bias of Schottky diode.

Voltage clamping

While standard silicon diodes have a forward voltage drop of about 0.6 volts and germanium diodes 0.3 volts, Schottky diodes’ voltage drop at forward biases of around 1 mA is in the range 0.15 V to 0.46 V (see the 1N5817[2] and 1N5711[3] datasheets found online at manufacturer's websites), which makes them useful in voltage clamping applications and prevention of transistor saturation. This is due to the higher current density in the Schottky diode.

 

[(*steve*)]

As for the circuit can someone explain why is it running in astable mode?

There are numerous answers:

1) Because that's the way you've wired it up.
2) because that's what you need in this circuit.
3) [insert explanation of how 555 operates]

 

[Notallbad]

Hi Could someone please kindly recommend which diodes to use in this circuit
Thankyou

 

[(*steve*)]

Go to your favourite component supplier and look up shottky diodes.

Almost anything should work, but I'd avoid large devices that attach to heatsinks. In fact, I might just go for the cheapest suitable component (suitable means packaging in this context),

 

[Notallbad]

Go to your favourite component supplier and look up shottky diodes.

Almost anything should work, but I'd avoid large devices that attach to heatsinks. In fact, I might just go for the cheapest suitable component (suitable means packaging in this context),

OK, thankyou I appreciate your help.
So I assume the 1A and 3A Schottky are overkill and the 0.2A would be suitable such as:
http://www.maplin.co.uk/schottky-rectifier-diode-19082

 

[(*steve*)]

I would imagine that would be fine. If the 1A diode is a lot cheaper, then it too may be OK.

Considering the cost, just thinking about this for a few minutes will eat up any difference

 

[KrisBlueNZ]

I doubt the circuit will work using a diode rectifier (D1). The AC voltage present at the capacitor under test will be very small, unless its ESR is very high; not enough to cause a diode with a 0.15V forward voltage to conduct. I would use an active rectifier based on an op-amp to detect the AC voltage on the capacitor.

 

[Jaynuts]

the diodes i used in the end were:

- UF4001
- BZX 85C I(IR 1)A6

the first one is shottky diode the second is zener diode. I can't remeber which values i used i dont have my old notes anymore. If i recall correctly one of them were 1.3A or something like that.

 

[Notallbad]

I doubt the circuit will work using a diode rectifier (D1). The AC voltage present at the capacitor under test will be very small, unless its ESR is very high; not enough to cause a diode with a 0.15V forward voltage to conduct. I would use an active rectifier based on an op-amp to detect the AC voltage on the capacitor.

Hi Thanks for your reply.

So my understanding would be that if the cap under test is 'faulty', i.e. has a high ESR, then the circuit would work indicating a high ESR reading, however if the cap under test is within tolerance then the circuit would fail and not give any indication of being 'Good'.
As I am only really interested in finding faulty caps and not in any way interested in good caps or what their ESR reading is then this circuit would suffice for my purposes.
OK, in that case its not an ideal working circuit but useful for my purposes at the moment.
I am in the process of gathering the parts needed to build the circuit on a breadboard and once built I shall report my findings/experience.
It should be apparent that my knowledge is limited and any help identifying the parts list to build the circuit would be helpful and very much appreciated.
I have a breadboard, hook up wires and a 555 on its way in the post, so now looking into which diodes, resistors and caps to use.
Thanks again.

 

[Notallbad]

the diodes i used in the end were:

- UF4001
- BZX 85C I(IR 1)A6

the first one is shottky diode the second is zener diode. I can't remeber which values i used i dont have my old notes anymore. If i recall correctly one of them were 1.3A or something like that.

Thanks, I can easily find the Schottky UF4001 but what voltage would I use for the other, BZX85C?

How did it perform after getting it working? Did it read good and bad caps, low and high ESR. It is also designed for in-circuit testing, did you ever try it out and how did it perform?

According to the original design authors notes he had it connected to a multimeter to take the readings, did you ever do this or just used an analog meter?

Thanks

 

[duke37]

The UF4001 is not a Schottky rectifier, it is a standard fast rectifier. A Schottky would have a lower turn on voltage.

 

[Jaynuts]

Thanks, I can easily find the Schottky UF4001 but what voltage would I use for the other, BZX85C?

How did it perform after getting it working? Did it read good and bad caps, low and high ESR. It is also designed for in-circuit testing, did you ever try it out and how did it perform?

According to the original design authors notes he had it connected to a multimeter to take the readings, did you ever do this or just used an analog meter?

Thanks

It performed pretty well, although i'd recommend high precision analogue meter. The one i had was crappy. Also I had no idea how to read the scale and apply it into resistance so the meter would only effectively show if the capacitor is broken or not. For exact results you will need to get two multimeters and get two readings to work out exact resistance.

 

[Notallbad]

It performed pretty well, although i'd recommend high precision analogue meter. The one i had was crappy. Also I had no idea how to read the scale and apply it into resistance so the meter would only effectively show if the capacitor is broken or not. For exact results you will need to get two multimeters and get two readings to work out exact resistance.

I appreciate your response, it sounds like it is as I suspected and just give an indication of good or bad which is all I really need right now (and probably all I ever would need) and cost is a key factor too so do not want to be spending much on something that is merely of personal interest - trying to repair a TV.

 

[Notallbad]

The UF4001 is not a Schottky rectifier, it is a standard fast rectifier. A Schottky would have a lower turn on voltage.

Well spotted, thanks for pointing that out!

 

[KrisBlueNZ]

So my understanding would be that if the cap under test is 'faulty', i.e. has a high ESR, then the circuit would work indicating a high ESR reading, however if the cap under test is within tolerance then the circuit would fail and not give any indication of being 'Good'.

As I am only really interested in finding faulty caps and not in any way interested in good caps or what their ESR reading is then this circuit would suffice for my purposes.
OK, in that case its not an ideal working circuit but useful for my purposes at the moment.

Typical ESR values for aluminium electrolytics range from less than 0.1 ohms to more than 1 ohm. An electrolytic may fail and go out of spec without having a hugely high ESR, so to be useful, an ESR meter needs to be able to measure ESR down to less than 0.1 ohms. A typical silicon diode such as the UF4001 (fast recovery) starts to conduct around 0.6V (at room temperature). A Schottky diode will start to conduct at about half that voltage. So you need to produce at least 0.3V across an ESR of less than 0.1 ohms. This requires a current of AT LEAST 3 amps before you'll even get a reading on the meter, even if you use a Schottky diode.

The only sensible approach is to eliminate the forward voltage issue, by using a precision rectifier. See Wikipedia.

 

[Notallbad]

Typical ESR values for aluminium electrolytics range from less than 0.1 ohms to more than 1 ohm. An electrolytic may fail and go out of spec without having a hugely high ESR, so to be useful, an ESR meter needs to be able to measure ESR down to less than 0.1 ohms. A typical silicon diode such as the UF4001 (fast recovery) starts to conduct around 0.6V (at room temperature). A Schottky diode will start to conduct at about half that voltage. So you need to produce at least 0.3V across an ESR of less than 0.1 ohms. This requires a current of AT LEAST 3 amps before you'll even get a reading on the meter, even if you use a Schottky diode.

The only sensible approach is to eliminate the forward voltage issue, by using a precision rectifier. See Wikipedia.

How might that be incorporated into the circuit design and using what op-amp?
Thanks

 

[KrisBlueNZ]

How might that be incorporated into the circuit design and using what op-amp?

Replace the circuitry to the right of the device under test with a half-wave or full-wave precision rectifier. See http://sound.westhost.com/appnotes/an001.htm (which is linked from the Wikipedia article on precision rectifiers) for a survey of possible circuits with advantages and disadvantages of each. You're using a relatively high frequency (well, ultrasonic is high compared to typical applications of a precision rectifier), so use a design with good frequency response. The diagrams in that article all use split supplies for the op-amps but you should be able to get away with a single supply if you use suitable devices.

A quick look on Digikey shows what should be a good choice, the Texas Instruments TLE2141A. It's also available as a TLE2142A (dual) and TLE2144A (quad). Current consumption is fairly high for battery operation, but it looks well suited to your application. The data sheet is sparse on application information but it does show how to null the input offset, which is probably worth doing in your application. It's also nice and cheap.

Edit: only the single op-amp version, TLE2141A, has the pins for offset null.