Is it possible to get two outputs (-5V and 0V [GND]) using a TTL signal?

I was thinking maybe I can use two comparators. When the TTL signal goes high I get -5V at one output A and 0V at the other output B. When the TTL signal goes low I get 0V at output A and -5V at output B. Not sure how to proceed though.

Any help would be greatly appreciated.

Question: Can a comparator have -5V and GND as it's supply voltages for V- & V+ respectively.

 

Yes. -5V isn't a TTL voltage though, so you are wanting something else.

I have a decent two transistor circuit that does the job. Want me to look it up?

Define your specs a little tighter though. TTL, for example, is typically +3.5V for a high and just under +0.7V for a low. What constitutes a high and a low for your new system?

 

Yes. -5V isn't a TTL voltage though, so you are wanting something else.

I have a decent two transistor circuit that does the job. Want me to look it up?

Define your specs a little tighter though. TTL, for example, is typically +3.5V for a high and just under +0.7V for a low. What constitutes a high and a low for your new system?

Hi Bill,

In my system a high is 5V and a low is essentially 0V. If you could look that up it would be great. Below is what I desire of a circuit.

At 5V TTL
Output A: -5V
Output B: 0V

At 0V TTL
Output A: 0V
Output B: -5V

I was thinking about using a comparator with GND at the positive supply and -5V at the negative supply. Is this feasible?

 

You might use a TTL to ECL level translator if you need to convert more than one signal, or have a need for high speed.

Here's an IC that has four inputs/outputs:
http://www.onsemi.com/PowerSolutions/product.do?id=MC10H424PG

 

TTL (HCT, ACT, BCT, FCT to name only a few types) is positive logic where <0.8V is a low and >2.0V is a high, there are no negative voltages in TTL logic.

You would need to generate a negative -5V supply (in addition to the standard +5V supply), and use transistors to do a level shift.

 

You might use a TTL to ECL level translator if you need to convert more than one signal, or have a need for high speed.

Here's an IC that has four inputs/outputs:
http://www.onsemi.com/PowerSolutions/product.do?id=MC10H424PG

It is only one TTL signal. Using this signal I need to have two outputs that switch using that TTL signal. Hopefully this makes sense.

When the TTL goes high the two outputs have the above voltages, when the go low the swap voltages.

 

Hello,

To have a better understanding on the logic levels, take a look at this page:
http://www.interfacebus.com/voltage_threshold.html

There is also a page for level translation over there:
http://www.interfacebus.com/Design_Translation.html

Bertus

 

TTL (HCT, ACT, BCT, FCT to name only a few types) is positive logic where <0.8V is a low and >2.0V is a high, there are no negative voltages in TTL logic.

You would need to generate a negative -5V supply (in addition to the standard +5V supply), and use transistors to do a level shift.

This is exactly what I want to do. Go from TTL to -5V and GND with a short propagation delay.

I had a circuit that did that using this part:
Link: http://www.nxp.com/documents/data_sheet/74HC_HCT04.pdf

but the propagation delay is too long. So I need help doing it differently. With transistors, comparator.

Need help.

 

What sort of loads will the outputs be driving?
How much delay can you tolerate?
Are your "TTL" levels coming from CMOS, or actual TTL logic? A TTL output does not go all the way to +5V.

 

What sort of loads will the outputs be driving?
How much delay can you tolerate?
Are your "TTL" levels coming from CMOS, or actual TTL logic? A TTL output does not go all the way to +5V.

The TTL is coming from an actually TTL logic.

The 0V & -5V will be driving a SPDT switch. So hence I need a driver circuit. The original driver circuit using the IC in my last post introduced too much delay.

From the TTL to the actually switching of the switch it has to be less than 5ns. The switch switches in less than 1ns. So essentially I have a max delay of approx. 4 ns.

Any ideas that can help me.

 

OK, the version I had was a negative to positive converter, so I redrew it. It was originally meant to do a negative logic system to printer interface, part of a light tree I had designed. The folks kept shifting the specs under me. I hate it when they do that.

A simple diode on an op amp/comparator would do the same thing.

I didn't do the two outputs because that is really basic, I figure you can take it from there.

 

The TTL is coming from an actually TTL logic.

The 0V & -5V will be driving a SPDT switch. So hence I need a driver circuit. The original driver circuit using the IC in my last post introduced too much delay.

From the TTL to the actually switching of the switch it has to be less than 5ns. The switch switches in less than 1ns. So essentially I have a max delay of approx. 4 ns.

Any ideas that can help me.

I'm skeptical. What kind of SPDT switch switches in 1nS?

 

CMOS switch?

 

CMOS switch?

Yeah, I guess it could be, considering that he wants complementary drive signals.

 

I'm skeptical. What kind of SPDT switch switches in 1nS?

It is a Triquint FET Switch.

 

I've got a circuit for you, but I have a couple of questions:
What is the risetime of your TTL input (what logic family are you using)?
What is the input capacitance of your Triquint switch? Can you post a link to the datasheet?

 

OK, the version I had was a negative to positive converter, so I redrew it. It was originally meant to do a negative logic system to printer interface, part of a light tree I had designed. The folks kept shifting the specs under me. I hate it when they do that.

A simple diode on an op amp/comparator would do the same thing.

I didn't do the two outputs because that is really basic, I figure you can take it from there.

Is there any way for this to switch faster? I used this transiter circuit and it inverted in less than 5ns but when the TTL comes back down it has trouble going back up. Link: http://www.play-hookey.com/digital/experiments/ttl_inverter.html

 

I've got a circuit for you, but I have a couple of questions:
What is the risetime of your TTL input (what logic family are you using)?
What is the input capacitance of your Triquint switch? Can you post a link to the datasheet?

The rise time of the TTL input is 2.5ns (wasn't given the class). The switch is at the link below. I don't know the capacitance but the isolation is a little over 30dB. Will be like 40dB at the frequency of interest, i.e. 300MHz.

Link: http://www.triquint.com/prodserv/more_info/proddisp.aspx?prod_id=TGS2306

 

Any ideas around a comparator with complimentary outputs?

 

1 ns is a very small figure, if it were a freq it would 1Ghz. I suspect even a comparator is going to have problems with that number.

It is in the realm of ECL, not TTL. ECL is a non-saturation logic, its logic levels are -0.7 and -2.0V, the basic circuitry is more like that of a differential amplifier. Their are TTL/ECL (and visa versa) converter chips out there.

And yes, ECL can come with complementary outputs, since it is a kind of differential amplifier. Speaking of which, have you looked into making a differential amp?