CMOS 4049-like Buffer with Inline Pinning

[Hi-Q]

Hey,

I am searching for a CMOS buffer (inverting or not) that accepts 15V suply, LVCMOS or TTL input levels (threshold around 1.5 to 2V), just like the CD4049, MC14049 and so on, but with layout friendly pinning (inputs on one side of the package, outputs on the other side).

I'd take tri-state buffers, bi-directional or whatever as long I can disable those special stuff. I just need something that does not require some crazy layout if I arrive with a parallal bus at the IC and want to leave it with parallel traces.

Regards,
Hi-Q

 

[barry]

I'm not aware of any 15V logic other than ancient CMOS. If you've got TTL or LVCMOS levels, why do you want to run your logic on 15V?
Maybe you need a voltage regulator.

 

[KlausST]

Hi,

Frequency, current, how many channels .....

Klaus

 

[Hi-Q]

I have to drive something that needs the 15V. While there are buffers with high voltage tolerant open collector or open drain outputs, this will have either slow rise times if I use high value pull-up resistors or high current with low state outputs if I use lower value pull-up resistors.

Frequency can be up to about 1 MHz. During switching the current is not that important, but during static times the quiescent current should be in the some µA range.
6 channels would be good, 8 would be nice, but I could live with 4 per package.

The CD4049/MC14049 I mentioned in the original post was the wrong number. The part that I wanted to refer to is the CD4504/MC14504. In CMOS mode it behaves like a CMOS device (virtually no quiescent current) and can translate 3.3V or 5V CMOS levels up to 18V levels.
The only drawback is the pinning where the inputs and outputs are on adjacent pins, which makes layout quite nasty. I'd like to keep the digital signals on one layer as much as possible in order to not take the risk of injecting digital spikes into the analog signals.

The LTC1045 has a nice pinning, but the quiescent current is 2.5 to 4.5mA and it costs almost $ 3, and it's a bit overkill with the latch and the setpoint referenc current generator.

 

[KlausST]

Hi,

What about CMOS analog switches SPDT.
Maybe not a low pincount solution...

Klaus

 

[barry]

I have to drive something that needs the 15V. While there are buffers with high voltage tolerant open collector or open drain outputs, this will have either slow rise times if I use high value pull-up resistors or high current with low state outputs if I use lower value pull-up resistors.

Frequency can be up to about 1 MHz. During switching the current is not that important, but during static times the quiescent current should be in the some µA range.
6 channels would be good, 8 would be nice, but I could live with 4 per package.

The CD4049/MC14049 I mentioned in the original post was the wrong number. The part that I wanted to refer to is the CD4504/MC14504. In CMOS mode it behaves like a CMOS device (virtually no quiescent current) and can translate 3.3V or 5V CMOS levels up to 18V levels.
The only drawback is the pinning where the inputs and outputs are on adjacent pins, which makes layout quite nasty. I'd like to keep the digital signals on one layer as much as possible in order to not take the risk of injecting digital spikes into the analog signals.

The LTC1045 has a nice pinning, but the quiescent current is 2.5 to 4.5mA and it costs almost $ 3, and it's a bit overkill with the latch and the setpoint referenc current generator.

If routing and noise is that big an issue for you, then why not use a multi-layer pcb? Why is slow rise time an issue; what is your requirement? You don't give us a lot of information. You're asking for a device with a TTL digital input and 15V output, and then you're worried about some analog signals; where does that fit in?

 

[Hi-Q]

If routing and noise is that big an issue for you, then why not use a multi-layer pcb?

It is a multi-layer pcb, but every added layer increases the cost even further.
Also the vias pollute other layers unless I order blind or burried vias and that gets really expensive.

Why is slow rise time an issue; what is your requirement?

I need up to 1MHz, so rise and fall times should be below 250ns

You don't give us a lot of information. You're asking for a device with a TTL digital input and 15V output, and then you're worried about some analog signals

Sorry, TTL was misleading. The digital signals are from 3.3V or 5V CMOS, so TTL threshold levels of about 1.5V would match the requirement.

; where does that fit in?

It fits in the realm of mixed signal circuits. The fact that the circuits process a mix of analog and digital signals does not have to mean that one wants to mix them together involuntarily.

 

[schmitt trigger]

The only other logic that I know which employed 15V supplies was HTL (High Threshold Logic) but that was even more ancient than CD40xx CMOS.

 

[barry]

It is a multi-layer pcb, but every added layer increases the cost even further.
Also the vias pollute other layers unless I order blind or burried vias and that gets really expensive.I need up to 1MHz, so rise and fall times should be below 250nsSorry, TTL was misleading. The digital signals are from 3.3V or 5V CMOS, so TTL threshold levels of about 1.5V would match the requirement.

It fits in the realm of mixed signal circuits. The fact that the circuits process a mix of analog and digital signals does not have to mean that one wants to mix them together involuntarily.

If it's multlayer, then put your analog on one layer and your digital on another. It sounds like you're getting really carried away when you start talking about blind and buried vias. Vias "polluting" other layers??? What have you got, a 128-bit ADC? Vias are used all the time in very high precision designs; what's so special about your design that it can't tolerate vias?

 

[Hi-Q]

Look, I just thought there might be an affordable IC out there that was similar to the MC14504 but with a more streamlined pinning.
If someone knows of such an IC I am happy to hear about it. That's all.