Inverter Nand Not Works

[ClassD2008]

Please help-me for my question.
Iam developmente test in MultiSim for Switing Power Suplie ( DAmp ).
But CD4070 NAND not inverter PWM Square Signal...
Seek in my circuit attached.

Please... help-me.

[RnDMonkey]

It might work better if you tried NAND gates instead of XOR gates. 😉 That would be the 4011 IC. 🙂

Also, do you really want that -55V on that ground? I guess I don't know your circuit so I can't say, but maybe if you fix those gates... 🙂

Message Edited by morgol on 05-30-2008 07:44 AM

[ClassD2008]

Thank you...
But really I need to use -55V and not ground.
In my circuit I use the CD4070.
You it tested my circuit attached?
Help-me.. please.

[RnDMonkey]

Well, your theory is correct but unless I'm mistaken, Multisim cannot handle an offset ground like that. Provided you make VSS for the 4070 chips be the -55 V and set their VDD 12V above that ( -43 V), the logic should come out right. The thing that is correct is that your top input to the first XOR is 12V above the -55V reference, and bottom input from off of Q1's collector is swinging up from -55V to 12V above that, relatively speaking. The problem is with the ground reference that the parts' logic use, so if you want to simulate it properly, you would have to move your gate input to the top of Q1 and bypass the 55V source. Alternatively, what MIGHT work is to build a 4070 part from the discrete transistors and such that make it, so you can force it to take your offset ground.

[ClassD2008]

I made one another test removing -VCC and placing Ground…
But exactly thus it does not works.
It helps.

[lacy]

I got the orinial to work and here my findings:

1). Changed -55V to VSS (not ground as these chips require VSS) and also add VDD as a precaution. Ryan is corrrect about this. The digital chips will only accept VDD, VCC, DGND or VSS and they can't be anegative voltage. Circuit still would not work

2). Examined the input to the first XOR Gate and saw a 1VDC square wave. The ouput was always high. This led me to believe that there wasn't enough signal amplitude for it to switch the XOR Gate. It was treating this 1V square wave as a low not matter what and this resulted in the ouput maintaining a high level (0+1=1 in an XOR Gate). Why wasn't the amplitude high enough?

3). Noticed that the transistor was a common base configuration. In this configuration you are not going to get any signal amplification and actually the gain is less than 1. So I rearranged the configuration to a common emitter ( or in this case a DC Switcher) configuration. That seem to rectify the situation

I will post this for you to examine. It may not be what you want but is the only way I could get it functioning.

[RnDMonkey]

Unfortunately, I had arrived at the same solution by switching to a common collector configuration, except I just replaced -55V with ground and tied the collector of Q1 right to ground to get the full pull down to 0V. Right now it should be switching from about 12V down to 6.8V, which should be in the right region that these gates would work (the model is based on 10V logic, I think). Unfortunately, this still doesn't solve his -55V problem. My suggestion is just to simulate it like this, and then intelligently add the offset ground as I said in my previous post. Rely on the simulator to do the grunt work, then rely on yourself to make up for it's shortcomings. And let's face it: Multisim is full of shortcomings.

[lacy]

Whoops, you are correct about that being a common collector and not a common emitter. I was thinking NPN transistor instead of a PNP. Everything is "backwards" with a PNP including the configurations. Anyway, reguardless of the type of transistor used, the way it is wired now is what I would refer to as a DC Switching Configuration (this may not be correct terminology, but that is what I have always called it). Well the good part is the circuit is working now and that is want we want.

[lacy]

Ryan, next time it might be a good idea to post the circuit file with the correction that you made. This way it will prevent anyone from duplicating your conclusions. I have always said a picture is woth a thousand words and it is really difficult on the forum to read a post and  try to visualize it in your head. That is why I always try to include a corrected circuit for the user to examine because I couldn't explain my way out of a paper bag most of the time.

[RnDMonkey]

I thought about it, but since I never actually solved his problem, I didn't think I should upload what I did. I guess I didn't want to upload a schematic without having the solution, since I didn't think that would be helpful. My mistake.