Get practical crystal tester working

Question

After working with quartz components in circuits, I planned to make a test on them in order to be sure if they are really still working as they are supposed to be.

I found this circuit on the internet:

I understand that the trick here is to block the DC current by the capacitors around and if there is an AC current flowing, which is coming from working quartz through the capacitors, then the LED will glow. That means the quartz component is working. If the LED does not glow, the quartz is not working because it is not able to produce AC current.

I thought in practical sense to make my own design for quartz testing, which I have tested in LTspice as working circuit. It is just below.

It works like the first circuit above which I have found on the internet. I soldered the components together to see what the result is going to be, but the result was negative. The quartz, especially the new ones, do not work. I was so frustrated.

Can you tell me what is wrong with the circuit I built?

Edit:

Here is the LTspice file:

Version 4
SHEET 1 880 1000
WIRE -1088 560 -1248 560
WIRE -992 560 -1088 560
WIRE -848 560 -928 560
WIRE -752 560 -848 560
WIRE -1248 576 -1248 560
WIRE -752 576 -752 560
WIRE -1088 608 -1088 560
WIRE -752 640 -848 640
WIRE -752 736 -752 640
WIRE -1088 752 -1088 688
WIRE -752 832 -752 816
WIRE -1088 976 -1088 816
WIRE -752 976 -752 912
WIRE -752 976 -1088 976
FLAG -1248 576 0
SYMBOL SBORKA\\QUARTZ_CRYSTAL\\XTAL_param -768 576 R0
WINDOW 39 -115 77 Left 0
SYMATTR InstName U1
SYMBOL voltage -752 816 R0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V1
SYMATTR Value 5
SYMBOL res -864 544 R0
SYMATTR InstName R1
SYMATTR Value 47k
SYMBOL res -768 720 R0
SYMATTR InstName R2
SYMATTR Value 1k
SYMBOL cap -1104 752 R0
SYMATTR InstName C1
SYMATTR Value 220p
SYMBOL cap -928 544 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C2
SYMATTR Value 220p
SYMBOL res -1072 704 R180
WINDOW 0 36 76 Left 2
WINDOW 3 36 40 Left 2
SYMATTR InstName R4
SYMATTR Value 100
TEXT -984 984 Left 2 !.tran 0.5ms
TEXT -1144 696 Left 2 ;Led

Answer 1

Can you tell me, what is wrong with the circuit, I had built together

A quartz crystal is a passive component and not an oscillator. You need to surround the quartz crystal with components that become a crystal oscillator i.e. it produces a sustained continuous sinewave of such and such a frequency. Your circuit applies a transient voltage of 5 volts that will cause the crystal to resonate but, that resonation will be very short lived and therefore won't illuminate the LED in a sustained manner.

In your top circuit, Q1, R1, R2, C1 and C2 form a Colpitts oscillator that sustainably "sings" at the frequency of the crystal (when it is attached).

Q2, R3, C3 and D1 is a type of AC magnitude detector that can measure the presence of an AC signal and take a proportionate current through the LED.

_{One thing to take note of is that the Colpitts circuit may oscillate when the leads are unconnected. Reason: the leads can form a transmission line that can look like an inductor and form a pseudo-standard Colpitts oscillator due to transmission line effects. This could lead to a false belief that the crystal is properly working when in fact it's broken and just behaving like a capacitor.}

Answer 2

The capacitors in the Colpitts oscillator circuit shown are relatively high value and maybe not ideal for modern many-MHz crystals. If you replace them with the crystal tester oscillator circuit typically used in the inexpensive frequency counter modules it may work better with high-frequency crystals:

^{simulate this circuit – Schematic created using CircuitLab}

The transistor shown (C9018) has a higher ft than the one in your schematic (1.1GHz vs. 600MHz) but either should work about as well for most crystals you are likely to find.

If you have a counter or oscilloscope or even an inexpensive DMM that has a frequency counter input, it is better to measure the actual frequency and compare with the markings. The frequency may be off slightly because of the load capacitance mismatch, but it should still be within a very small percentage.