[solved]74HC4053 output issue

[desai_amogh]

I am working on a sdr receiver with 74HC74 and 74HC4053 as a QSD mixer.  I built the circuit and 74hc74 does fine job by dividing input frequency by 4.  However, at 74HC4053 outputs I get the same divide by 4 at input and output. I have traced all the connections everything looks fine.  What could be the issue?

Please see the attached circuit.

So for example, at 10Mhz input frequency to 74HC74, Input at Pin 9 and 10 of 74HC4053 is 2.5 Mhz (divide by 4 from 74HC74).  However, at pin 1, 2, 3 and 5 of 74HC4053 I get same 2.5Mhz as output too.

[Andy Watson]

Charge injection - or similar switching artefact entering the signal path. See "crosstalk voltage" under "Additional dynamic characteristics" in the data sheet:
http://www.nxp.com/documents/data_sheet/74HC_HCT4053.pdf

[desai_amogh]

Charge injection - or similar switching artefact entering the signal path. See "crosstalk voltage" under "Additional dynamic characteristics" in the data sheet:
http://www.nxp.com/documents/data_sheet/74HC_HCT4053.pdf

Hi Andy,  Does it mean there is a same frequency signal entering from the VCC which I need to filter using a inductor or something?

[tggzzz]

Without an picture of the output, this is speculation.

The digital inputs are connected to the pass transistors. The pass transistors have parasitic capacitances between the various terminals. Voltage changes on the digital inputs will cause transient currents to flow in the parasitic capacitances, and will therefore be coupled to the output. Key parameter in such switches is the "charge injection", specified in pC.

Other effects: differing on-resistance and other threshold effects are different for each of the transistors, and will be visible as voltage pedestals on the output.

Other effects: poor layout separating the analogue and digital sections of the circuit.

[desai_amogh]

Without an picture of the output, this is speculation.

The digital inputs are connected to the pass transistors. The pass transistors have parasitic capacitances between the various terminals. Voltage changes on the digital inputs will cause transient currents to flow in the parasitic capacitances, and will therefore be coupled to the output. Key parameter in such switches is the "charge injection", specified in pC.

Other effects: differing on-resistance and other threshold effects are different for each of the transistors, and will be visible as voltage pedestals on the output.

OK, I guess I know the issue then.  I have designed a PCb for the above circuit using kicad. This is my 1st PCB design and I have tried to keep it small at 5cm x 10 Cms. It must the lay out issue as per this:

Other effects: poor layout separating the analogue and digital sections of the circuit.

[dannyf]

However, at pin 1, 2, 3 and 5 of 74HC4053 I get same 2.5Mhz as output too.

Without knowing how you measured it, it is hard to interpret that correctly.

[desai_amogh]

However, at pin 1, 2, 3 and 5 of 74HC4053 I get same 2.5Mhz as output too.

Without knowing how you measured it, it is hard to interpret that correctly.

I have a frequency counter.

[IconicPCB]

What do you get at line out?

[desai_amogh]

What do you get at line out?

I havent measured frequency at line out. but I guess the filter before the opamp will remove the HF frequencies.  But when i connect it to the computer sound card and run HDSDR, I dont get any signal. There is the thin noise line at the middle of bandscope and only at 1086Khz I get a local broadcast station. Whose image is also seen on the other side of the central noise line.  When i tune around, this broad cast station and its image also moves with tuning. I dont get any other signal at any other frequency with tuning range from 0 to around 8 Mhz (vfo upto 32mhz). Here is a image of it.



Note: Though you see it at 1020khz, the broadcast station is actually transmitted at 1086khz and my vfo might have a little offset.

[tggzzz]

Without an picture of the output, this is speculation.

The digital inputs are connected to the pass transistors. The pass transistors have parasitic capacitances between the various terminals. Voltage changes on the digital inputs will cause transient currents to flow in the parasitic capacitances, and will therefore be coupled to the output. Key parameter in such switches is the "charge injection", specified in pC.

Other effects: differing on-resistance and other threshold effects are different for each of the transistors, and will be visible as voltage pedestals on the output.

OK, I guess I know the issue then.  I have designed a PCb for the above circuit using kicad. This is my 1st PCB design and I have tried to keep it small at 5cm x 10 Cms. It must the lay out issue as per this:

Other effects: poor layout separating the analogue and digital sections of the circuit.

Before jumping to that conclusion, you really should examine various waveforms with an oscilloscope. It would be sensible to post a picture here for comment.

Given the effect that poor probing technique can have on measurements, it would be advisable to post a picture of your probing technique.

[desai_amogh]

Without an picture of the output, this is speculation.

The digital inputs are connected to the pass transistors. The pass transistors have parasitic capacitances between the various terminals. Voltage changes on the digital inputs will cause transient currents to flow in the parasitic capacitances, and will therefore be coupled to the output. Key parameter in such switches is the "charge injection", specified in pC.

Other effects: differing on-resistance and other threshold effects are different for each of the transistors, and will be visible as voltage pedestals on the output.

OK, I guess I know the issue then.  I have designed a PCb for the above circuit using kicad. This is my 1st PCB design and I have tried to keep it small at 5cm x 10 Cms. It must the lay out issue as per this:

Other effects: poor layout separating the analogue and digital sections of the circuit.

Before jumping to that conclusion, you really should examine various waveforms with an oscilloscope. It would be sensible to post a picture here for comment.

Given the effect that poor probing technique can have on measurements, it would be advisable to post a picture of your probing technique.

My issue is fixed. the inductor at pin 4/15 of 74hc4053 was short/bad, replaced it and sdr works!