Everything else is there except the maximum clock frequency. I'm designing a SSB transceiver with four lm386 and two cd4053 in the exciter section.

What's the maximum clock frequency of CD4053? Can anybody help and share from experience please. Does it work at 7MHz?

Please help!!

Since the CD4053 is an analog multiplexer, it doesn't actually have a clock. The analog switches are rated for 30 MHz.

Curtis Eickerman

Thanks for your reply. I'm trying to come up with an easy to build ssb transceiver with common parts. CD4053 is widely available all over the world. 74HC4053 the high speed version is only available from mouser india digikey india and few others. Attaching the schematic of product detector and modulator stage. One spdt is 4mhz oscillator and it's feeding another spdt (single balanced mixer).

Entire design is on paper. I've been dealing with few health issues. Do you think this would work?

But can they be switched that fast?

The issue I remember was that some analog switches were lossy even in the on position. That was at audio, but likely is a concern at radio.

74HC series are faster but only selected stores keep them. CD4053 is as common as housefly !

CD4066 darasheet says 8MHz max clock frequency. CD4053 must be similar

Lossy means Ron resistance. Per SPDT it’s 150ohm or less. Gain in amp stages make up for the losses.

I’ll upload an old article. The author used cd4066 to successfully generate 4MHz SSB. Of course a crystal ladder filter was there too

According to the Texas Instruments data sheet:
https://www.ti.com/lit/ds/symlink/cd4053b.pdf?ts=1613587625890&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FCD4053B
The address to signal out (channels on or off) can be anywhere from 120ns to 450ns (typical) and as much as 240ns to 720ns (maximum) depending on the supply voltage. See the bottom of page 7. I would say that you'd want to use a 15 volt supply voltage to have any hope of making the circuit work at 4MHz.

The CD74HC4053 would definitely work:
https://www.ti.com/lit/gpn/cd74hc4053

John

The schematic.

CD4066 should work better. Datasheet gives a typical frequency of 8MHz at 12V.

Very old article.. from 1988... 4.194MHz is easily handled by the CD4066. It is fed from an audio transformer

Thank you, John. CD4066 looks much better. Adding inverter gates to drive 4066 will increase parts count but performance is going to be excellent.

CD4066 is common as dirt and a decent switch if you can tolerate its high on resistance. Is crosstalk or capacitance feed-through an issue? There are lots and lots of analog switches. DG201-202 are pretty decent and should work well at 4mHz. I'm sure there are others.

Hi Richard,

CD4066 is better, reliable, and has channel isolation of -50dB which is good.
I've redesigned the circuit with 4066 and three of them would be required. I'm trying to draw the BFO around the two remaining switches so I don't have to use an inverter IC like 74HC04 to drive the CD4066 SBM mixer. To overcome losses I’ll see what I can do. I think Ron is 150ohms at 12V. Paralleling 3 would give me 50ohm. Here’s the rough schematic. Gain of LM386 is set to 20. The distortion and noise would be low enough for cascaded LM386 audio amplifier. A transistor preamp has also been added increasing the audio stage gain to 85dB (little excessive? )

Have you considered MC1496? I'm not an expert here and may actually reveal my lack of knowledge, but it seems tailor made to do what you want, check out figure 28 in this datasheet:
https://www.onsemi.com/pub/Collateral/MC1496-D.PDF

Do you really need so much audio gain, 85db? That would amplify a 1/2 millivolt signal to 10 volts, which would be crazy loud if it weren't more than the speaker's LM386 can deliver. A microphone close to your mouth would need only half that, I would guess depending on microphone and exciter needs. Noise and audio instability could be a concern with so much gain. Looks like a fun project, though. SSB walkie talkie!

CMOS switches as balanced mixers aren't a new thing, and they tend to emulate more traditional balanced mixers.

They have actually seen a lot if use in high end design.

Nowadays, CMOS switches are more available (is the 1496 still in production?) and likely cheaper than an analog mixwr.

For a lot of things your oscillator is putting out a square wave, so feeding switches is no problem.

And yes, I cringe at seeing a 386 being used in place of opamps.

I’ll take the suggestions and lower gain. LM386 is used for simplicity. Powering them with 6Vcc and 1K load would keep distortion at 0.2% . A transistor phase splitter then feeds the SBM.

Yes I’m aware of MC1496 but it’s not easily available. My design goals were compact IC design and eliminating toroids.

Here comes another design, is it too simple to work? The front end has a diode DBM.

I think I'm trying to make a bidirectional amplifier with J310 fets. With the gate grounded, it's common to both drain and source and signals should flow either way.
Is there any gain when drain (To gnd or -) and source (To +) are connected to opposite supply rails?

A third J310 RF preamp (12ma) may be placed before the DBM for added sensitivity.

Mic and speaker are connected to the product detector.

Source resistors are missing in J310 amps.

Wow I like your creativity but I don't see how this can work effectively in both signal directions. You may be way ahead of me but what I see is, for instance, once you've set bias to have gain in one direction, if you switch supply polarity, the device is suddenly at the opposite polarity (relative to the grounded gate). You've just switched the fet off for the most part. It may turn on briefly for any brief signal peaks that exceed the source voltage but not what I think you intend. What am I missing?

MC1496 surface mount variants are still in production. I'm not sure about the status of DIP variants but they appear easy to find here from surplus/overstock vendors.

You got a valid point Richard. Maybe someone else has done it too. I'll test it out on Spice if it has a model for J310. My feeling is both the coils have to be tuned at IF for better amplification. See this attachment this circuit is better. Two 1.3V zener diodes added for switching. It's the lowest voltage zener I could find. Only few online sites carry them.

Just watch out for the operating voltage between the 74HC and the CD.types. They are not readily interchangeable. The 74HC will only work up to 5 volt inputs whereas the CD type should work up to around 15 volt inputs.

I found this out earlier today when trying to sub a 74HC 4051 for a CD4051 ina temperature controller. The 74HC would not do the multiplexing with the 15 volt supply that was being supplied.

An early homemade transistor SSB transceiver just arranged things right.

A common filter and IF strip just going one way.

At one end, it was fed from a mixer, or balanced modulator, depending on receive or transmit.

At the other end of the IF strip, a balance.d diode mixer that worked as a product detector on receive, or a mixer on transmit.

It made everything very simple, rather than trying to make each stage work in two directions.But it's easy because of it being single conversion. The SBE-33 is tge classic example of bilateral stages, and maybe the ancestor of later work, but I'm not sure it really simplified anything.