I have a 10 MHZ time base circuit (see attached). It goes through a chain of 74LS90s to divide it down to 1 Hz.
It count up to 99999 just fine. When probed with my scope, I get messages such as "Sine Wave Detected on Ch1",
"DC Level Detected on Ch1", "Unknown Signal on Ch1" and sometimes "Square Wave or Pulse Detected on Ch1".
If I feed a 10 MHZ square wave into the counter chain with my function generator, there is a good clean square
when various points are probed.

Any suggestions would be appreciated.

Thanks

 

When probed with my scope, I get messages such as "Sine Wave Detected on Ch1",
"DC Level Detected on Ch1" and sometimes "Square Wave or Pulse Detected on Ch1". Any suggestions would be
appreciated.

Where are you probing? What is the supply voltage? What is the scope bandwidth? Are you using a 10X probe? Has the probe been compensated?

Circuit in question:

 

Where are you probing? What is the supply voltage? What is the scope bandwidth? Are you using a 10X probe? Has the probe been compensated?

Circuit in question:
View attachment 226188

Hi Dennis

probing the time base output and various points in divider chain
5 volt supply
scope bandwidth is 50 MHZ

Tektronix TPP0051 10X passive probe

Length:	1.3m
Applications:	Low-frequency measurements, Low-frequency computer and telecom measurements, Power supplies, Low-frequency amplifiers
Attenuation:	10X
Bandwidth:	50MHz
For Use With:	BNC oscilloscopes with 1 MΩ/20 pF inputs.
Input C in pF:	<12 pF
Input R in Ohms:	10 MΩ ±1.5%
Input Voltage:	300 VRMS CAT II
Risetime:	<7 ns

 

Can you share a picture of the actual parts unit please.

Oscillators are incredibly sensitive to how they are wired up,
the difference between one oscillating and one not can be just how a wire is placed,

What actual traces do you get on the scope ?

 

Your scope will only display the third harmonic well (the 5th will be 3dB down), so the corners will be rounded but it should still be more like a square wave than a sine wave.

Has the probe been compensated? The probe is also 50MHz, so it will give another 3dB attenuation at 50MHz.

Pictures of the waveform and messages would be helpful.

 

That schematic doesn't show the supply voltage.

Nor does it show where you are probing the circuit.

The 74HC04 output driving the crystal will presumably be
full gnd to Vdd which seems excessive for the crystal.

Also the 4.4K (two 2.2K resistors) across the crystal
isn't the environment the crystal needs for best
response.

I'd guess that you've actually built this:

 

That schematic doesn't show the supply voltage.

It's common for power supplies to not be shown. It removes unnecessary clutter. OP specified in post #3.

Your schematic doesn't show supply voltage either...

 

Your scope will only display the third harmonic well (the 5th will be 3dB down), so the corners will be rounded but it should still be more like a square wave than a sine wave.

Has the probe been compensated? The probe is also 50MHz, so it will give another 3dB attenuation at 50MHz.

Pictures of the waveform and messages would be helpful.

The messages are:
"Sine Wave Detected on Ch1", "Unknown Signal on Ch1", "DC Level Detected on Ch1" and sometimes
"Square Wave or Pulse Detected on Ch1".

When the message says square wave, the display shows a decent wave form with slightly
rounded corners. When any of the other messages appear, the wave form is a squiggly
line with lots of noise.

The probe is compensated.

I am wondering if the oscillator circuit is no good because a 10 MHZ square wave
from my arbitrary function generator results in perfect wave forms no matter where I place
the probe.

 

You need to show us your actual circuit ,
preferably with the scope probes attached, and a picture of the scope trace,
I'm sorry , but the error messages tell us very little,
a wiggly line, will tell us something,

Im also concerned about your circuit,
when I have made TTL based crystal oscillators, I have some capacitors in there, and the resistor is of the order of many M ohm,

e.g. https://www.electronics-tutorials.ws/oscillator/crystal.html

Your circuit also seems to have the crystal connected across a buffer, not an inverter,

this is more similar to your circuit, ( circuit 1d)
but note the capacitor in the middle
https://www.analog.com/media/en/technical-documentation/application-notes/an12fa.pdf

 

When the message says square wave, the display shows a decent wave form with slightly
rounded corners. When any of the other messages appear, the wave form is a squiggly
line with lots of noise.

You need to show us the waveforms that correspond to the messages. If it isn't clear, we also need to know the vertical sensitivity setting.

I am wondering if the oscillator circuit is no good because a 10 MHZ square wave
from my arbitrary function generator results in perfect wave forms no matter where I place
the probe.

Your function generator has more drive capability than a 74HC gate. Your oscillator circuit might be on the verge of instability and works when there's no extra loading, but adding the probe affects it. It would be informative to know what happens to the output when you get the questionable waveforms and messages.

 

You need to show us the waveforms that correspond to the messages. If it isn't clear, we also need to know the vertical sensitivity setting.
Your function generator has more drive capability than a 74HC gate. Your oscillator circuit might be on the verge of instability and works when there's no extra loading, but adding the probe affects it. It would be informative to know what happens to the output when you get the questionable waveforms and messages.

The output drops when I probe the output.
How can I increase the drive of the oscillator?

 

The output drops when I probe the output.

The output of the HC14 shouldn't be loaded by your scope probe. What is it driving?

How can I increase the drive of the oscillator?

If you have spare gates in the 74HC14 package, you can put some in parallel with the existing gate. This isn't recommended if the gates aren't in the same package.

 

The output of the HC14 shouldn't be loaded by your scope probe. What is it driving?
If you have spare gates in the 74HC14 package, you can put some in parallel with the existing gate. This isn't recommended if the gates aren't in the same package.

 

Its driving seven 74LS90s in a divider chain to get to 1 HZ and five 74LS90 / 74LS375 / 74LS47 counter/display stages and a dual 74LS123 that will control the store data, counter reset, and a 74LS14 for the power-up logic.

 

Its driving seven 74LS90s in a divider chain to get to 1 HZ and five 74LS90 / 74LS375 / 74LS47 counter/display stages and a dual 74LS123 that will control the store data, counter reset, and a 74LS14 for the power-up logic.

What is the HC14 actually driving? It can only be driving one of the LS90's.

 

That is all that it is directly connected to.

There shouldn't be a loading issue. 74HC can drive 10 LSTTL loads. Bus drivers have only half the strength of LSTTL drivers (15 loads vs 30).

 

Is there a better way to generate the 10 MHZ signal. Should I purchase a Saronix
10 MHZ oscillator module? https://datasheetspdf.com/datasheet/NCT050C.html

Or should I not use a CMOS chip for the oscillator?

This is disappointing as I am so close to getting this project done.

 

I've heard that using a packaged oscillator module is easy and works, while people have problems with other circuits.

Or you could try something like this (remember that at 10 MHz you need short wires).

 

The output of the HC14 shouldn't be loaded by your scope probe. What is it driving?
If you have spare gates in the 74HC14 package, you can put some in parallel with the existing gate. This isn't recommended if the gates aren't in the same package.

BIG NOTE:

paralleling outputs in same package is a good answer to increase drive,

BUT, you scope, assuming its has a probe , is NOT going to load the output of the 74HC14 to an extent that you would notice,
So paralleling up the gates is MASKING the REAL PROBLEM, making it harder for you to find it.

Find and fix the problem.



Second up, you have not provided actual pictures of the setup, this is most likely the reason for the problem,

Third, you are now asking about off the shelf oscillators,
If you don't want to make your own, then a much cheaper and reliable option is an off the shelf 10 MHz TTL oscillator.
almost no one makes there own 10 MHz oscillator now days.

If you want through hole,
an example
https://www.digikey.co.uk/product-detail/en/epson/SG-531P-10-0000MC-ROHS/SER1208-ND/1021904


or SMD, an example
https://www.digikey.co.uk/product-detail/en/ecs-inc/ECS-5032MV-100-BN-TR/XC3097CT-ND/9953277


or if you want REAL stability

https://www.digikey.co.uk/products/...t=1000011&page=1&nstock=1&rohs=1&pageSize=500

( just joking, just to show you what's around )

 

BIG NOTE:

paralleling outputs in same package is a good answer to increase drive,

BUT, you scope, assuming its has a probe , is NOT going to load the output of the 74HC14 to an extent that you would notice,
So paralleling up the gates is MASKING the REAL PROBLEM, making it harder for you to find it.

Find and fix the problem.



Second up, you have not provided actual pictures of the setup, this is most likely the reason for the problem,

Third, you are now asking about off the shelf oscillators,
If you don't want to make your own, then a much cheaper and reliable option is an off the shelf 10 MHz TTL oscillator.
almost no one makes there own 10 MHz oscillator now days.

If you want through hole,
an example
https://www.digikey.co.uk/product-detail/en/epson/SG-531P-10-0000MC-ROHS/SER1208-ND/1021904


or SMD, an example
https://www.digikey.co.uk/product-detail/en/ecs-inc/ECS-5032MV-100-BN-TR/XC3097CT-ND/9953277


or if you want REAL stability

https://www.digikey.co.uk/products/en/crystals-oscillators-resonators/oscillators/172?k=oscilator&k=&pkeyword=oscilator&sv=0&pv183=316150&sf=0&FV=69|411897,mu10MHz|2150|0,mu10MHz|2150|1,-8|172&quantity=&ColumnSort=1000011&page=1&nstock=1&rohs=1&pageSize=500

( just joking, just to show you what's around )

Here is my freq ref circuit and counter chain.