Bizzare EPROM behavior

Go To Last Post
21 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I've finally gotten my EPROMs programmed, but the output voltage of one of them is 3.5V on all DATA pins, while the other ones have normal output voltages (5V). The type of the first one is HN27256G-25 and the normal one is M27C256.

Why is this so? Is it a line of low voltage EPROMs?

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I think the 27256 is "NMOS" and the 27C256 is CMOS. They will, I think, have different output logic levels. CMOS generally pulls up to Vcc. IIRC, NMOS is more like TTL with mediocre pullup. These are ancient chips (esp 27256)!

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

So I'm supposed to add pullups? Like 10k? Please define mediocre :-)

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The pullups are in the chip. You shouldn't need to add pullups and 3.5 is well above the 2.0volt threshold for TTL. But what else is on your databus, TTL, CMOS?? Try not to mix family types on a common bus.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Well, in that part of the system, there's no bus. The output of the EPROM is just command lines. It's for the microcode.

The lines from that strange one go on these:
74LS138 //3 lines
CD74HC151 //3 lines
74HCT193 //2 lines

The OE is set to zero, zeroes are there where they are supposed to be ( a nice 0.05V), but the 3.5V is strange.

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

Last Edited: Tue. Jun 5, 2007 - 06:35 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Is there data on the lines while you are measuring this voltage?

Go electric!
Happy electric car owner / builder

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

sgomes: Yup. The data is correct, the zeroes and ones are in place, but the ones are low voltage ones.

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You are looking at it with a scope right? What happens, When you disconnect the data lines from their destination (AVR I assume). So EEPROM data outs are sending data to air particles not AVR. is the voltage correct then?
Can you read the data with the AVR? is anything getting hotter than it should be? Does the datasheet say anything about it?

Lots of questions, but probably easy and quick answers. take it as a checklist.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

So:

The target is NOT an AVR. There is NO AVR anywhere.
Not EEPROM but EPROM.
When disconnected, the voltage is still bad.
There's no extra heat. Just the warmth you get from any LS chip :-D
If I had an AVR, I cuold read it, because the programmer at a company I visited programmed it, verified it and was happy.

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

daqq wrote:
Well, in that part of the system, there's no bus. The output of the EPROM is just command lines. It's for the microcode.

The lines from that strange one go on these:
74LS138 //3 lines
CD74HC151 //3 lines
74HCT193 //2 lines

The OE is set to zero, zeroes are there where they are supposed to be ( a nice 0.05V), but the 3.5V is strange.

Lot's of family types there, your mixing TTL with CMOS. If this whole thing runs slow then you might not have noise/overshoot/undershoot/race problems.

How fast is this circuit being clocked? Are you seeing operating problems or just concerned about that one chips output voltage?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I haven't ran it yet. But it will be clocked somewhere within 10hz (for debugging purposes only) to about 100K - 1MHz (to see how fast the bugger can go without errors).

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Daqq -

Do not worry that one has 3.5V output. I think that if you look at its data sheet, you will find that it is correct for that chip. That is the way it is. Live with it or change it.

Now, SOME logic will be fine with it (like TTL). Some (like some CMOS) won't. If it is an issue, you wil need to go through your system and do a signal level analysis. Its what any designer of a mixed-logic system has to do.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Daqq,

What you have is an generic 250nS, 27256 (32K Byte) NMOS EPROM.

I have attached a generic datasheet.

In that datasheet, the minimum voltage for a legal logic high is 2.5 volts.

The 3.5 volts that you are seeing is pretty much mormal for the older (non-CMOS) EPROMS.

If you don't like the 3.5 volts as opposed to one that outputs 4.5 volts or greater, use a CMOS EPROM. The 3.5 volt output is typical for NMOS technology. Back in the late 1970s & 1980s, we would think nothing of an EPROM whose outputs were 3.5 volts. NMOS EPROM technology was designed to be compatible with TTL technology. It should cause no problems if used with CMOS logic, if the data signals don't excede 8MHz to 10MHz.

Attachment(s): 

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks you all for solving my mystery. I never dealt with EPROMs before, so I was a bit surprised thatt it gave 3.5V. I was just a bit surprised, thinking that it had gone bad or that I had a short.

I was under the impression that the old tech used good old TTL values.

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

daqq wrote:
I was under the impression that the old tech used good old TTL values.

That is a good old TTL value. After spending many years on a bench troubleshooting TTL logic, I'd say that 3.5 volts at a TTL output is pretty typical for pre-74Lxx/74LSxx/74Sxx technology.

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I agree with Carl. Good old TTL. 3.5V is about right for logic high.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kid now-a-days! The "New Fangled" stuff they think is from the "Good Old Days!" My word!!!

Lets talk about RTL, DTL, ECL, Descrete Diode Logic, Descrete Transistor AND/NAND/OR/NOR/XOR/XNOR gates, Flip-Flops, etc... And then, I suppose, Vacumm Tube digital logic will really be pushing us into the Museum!!!

Actually, I have some ECL here, at the House! I've been itching for a reason to use it for about 25 years. Haven't found one yet! I don't think it's even used anymore. Marvelous stuff! In its day, ECL was refered to as, "Analog Digital."

That's because ECL didn't operate its transistors in saturation, like all of the other logic families. ECL was also known as "Linear Digital..." The ECL designer also needed to understand transmission line theory too!

And you talk about a "Gas Hog!!!" That would be ECL...

Kid now-a-days!

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes, I cut my teeth on RTL (the REAL RTL, not this new-fangled Register Transfer Logic), and ECL.

In grad school, an instructor required us to design an ALU in TTL (this was after 8008 microprocessors had been out several years). He docked us because it wasn't designed to industrial standards. I complained that "nobody" would design an ALU using plain TTL, let alone to "industrial standards" (this argument ignored what NASA was doing, of course, but that wasn't "industrial").

Oh, well, lots of stuff under the bridge since then.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:
In grad school, an instructor required us to design an ALU in TTL...

He docked us because it wasn't designed to industrial standards. I complained that "nobody would design an ALU using plain TTL, let alone to "industrial standards"...

Noboldy, except Daqq.

Daqq, your TTL controller will be to "Industrial" standards, won't it? Else yoou'll be docked points! :lol:

Oh! There will be points decucted too, if the final design has a power supply that is out of spec! :roll:

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Nobody, except Daqq and me :D except that mine is all HC cmos ttl equivalents.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
except that mine is all HC cmos ttl equivalents.

Well, I used LS and that because I had them from some device I dissasembled not long ago. And because I am a bit broke :-(

carl: Heh, I was actually thinking of using DTL as a data comparator (would determine if the data on the bus is equal to zero), but then I used a bit of a hack: Pulldown resistors on the bus. Thanks to it I can clear the whole 32k of memory with a single instruction that takes about 65536 clock cycles. Possibly the longes instruction ever :-D . Long live(and process :-D ) the microcode!

Due to that I was not sure what DTL would do, so I rather used 7 OR gates. I also wanted to use one NAND (composed of discretes) instead of a whole 7400 (I only use one for blocking the clock when an instruction is being read), but I though that uniformity was good. And in the end it would most likely be larger than a 7400.

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.