_q12x_

DIY
Thanks to mister @marconi , I received his chip AT28C16 gift from England !
I will start to make the 5x7 led matrix project that you did.
I will make it alone as much as I can (as usual). And I will ask you for clarifications on the way.
 
That is good news. You will find this eeprom easy to program and use. It has enough memory for it to contain several sets of code each set being for its own project. Just move the chip from one board to another and arrange for each board to address its set of code. The eeprom has 2exp11 x eight bit memory locations.

The first task is to construct a programmer similar to the one I showed you on YouTube.
 
Hello mister @marconi
Just today I stay over the circuit from Ben Eater , and I made it and it works ! Youhoo.
Im happy because it didnt work from the first try. And I had to spend some hours looking like an idiot to the absolutely correct wiring and not understanding why is not working. I solved the --blody-- puzzle realizing that the LED's I was using were the SuperBright White ones. So in the moment I switch them with an older type of colored LED, everything worked like a charm. I have the exact same circuit as Ben has in its video. What a road... uuuh. I thought for a momment I am completely stupid.
 
Yesterday I stay and made the actual circuit. Here it is for others to see/use:
ATMEL AT28C16 noSim-circuit only - Copy.jpg
 
I will start to use the LED Matrix later today.
I will probably stay and draw the special pins this components has and the proper wiring... hmmm
or more experimentation. It will take time to do it alone. But I will do it.
 
I wonder what EEPROMs are out there, SMD/DIP versions, the most popular, the CHEAPEST and the more quantity of them that can be bought.
- I want 100 EEPROMs, for under 10$ - doesnt matter from what e-market or what size/package they have. Haha. But seriously, give it a look.
 
I found it useful to practice programming the eeprom in order to learn the routine. I made mistakes some of which I thought would damage the input/output circuits but they seem quite robust though I would not rely on that.
 
I meant why not use the proteus circuit as your blueprint to build it in reality.

It is up to you but there are fewer programming iterations for the letters, numbers and symbols you wish to display if you scan column by column ie you will programme for each 5 memory locations whereas if you scan row/line by row/line you will have to programme 7 memory locations.
 
I see. I am using my proteus circuit that I built already, ofcourse but I used some other memory chip in there, if you remember. Ill have to adapt to this one here. It was good if you had already the circuit prepared and ready, but is fine without it. I will manage with what I have.
I had to do a lot of preparations, download pdfs, rename a lot of IC's pdfs in my PC, I had to make a new order in everything. And then retrieve the pinout for each IC, print them, and now hopefully assemble them all correctly.
In my original circuit simulation I used a generic FlipFlop with a D linked to /Q.
But now I am using 4027 which have K an J. Which one I should use ?
Memory and led matrix display v7g for print seg1.jpg
 
I built separately another circuit for 7493, (that counts to it's maximum 16 beat's)
1663591689500.png

with the same IC as before ,but also with another brand new, and I got the same random result, so the IC is good.
I also replace that big button with a brand New button and again, some slight improvement, But still not good enough.
I built in the movie, a button debouncer using 470nF, After movie, I switch to a 10uF and now to a 200uF. The difference is visible and better but still not good enough. I believe I might have to make a button debouncer using a 555.
I strongly believe is the button fault.
 
The push button de bounce circuit can be just like the one for the eeprom programmer. This circuit produces a nice sharp negative going pulse when the button is pressed. The 7493 has clock inputs which are triggered by negative going transitions ie 5V to 0V.
 
Is working now...with that 555 debounce circuit...what a day !
I really wonder what's your trick that you can figure out so quickly the sequence of the 7493? And not using a simulator like I do. If you can explain your process in words will be more than good.
You are right, indeed it counted LESS than 7, now that I can actually have a clear button press, without any monkey bizniz. I know in my simulation I did it right.. but I have to recheck it now.
 
Home - BBC News - https://www.bbc.co.uk/news

As friends of the United Kingdom I thought you and Romania might like to look at some of the television coverage of the funeral of our late dear Queen Elizabeth. Most of today many (millions!) have been watching the television and listening to the commentaries, music and prayers - very moving and a glorious way to mark Her Majesty's passing. I hope you gain a sense of how loved and admired she was by ' the great and the low'
 
Last edited:
  • I was having today a strange feeling about UK queen funeral... without watching any news, because I was working all day long like a slave. I had a brief feeling at some point. Strange things are transmitted, even feelings.
  • Indeed, Romania is transmitting to England it's sincere condolence and we are sharing a tear along with you with a piece of broken hearth as well.
 
1. Connect a 0.01uF capacitor between 5 and 1.
2. Connect a 0.1uF in parallel with a 1uF electrolytic between 8 and 1.
Is the pushbutton 'normally closed' as you have drawn?
The button is staying in a open position and is closed/contacting when is pressed.
This is the type of button I am using:
1663664584070.png

Im sorry Im not that bright as you may think I am, what 5 and 1? and what 8 and 1 are you referring to?
----
Ive tested separately the 7493 and it turns out is a good chip, undamaged !!! GOOD !
Then it remains theNAND and theAND chips to be verified. And if they are bad, I will have to build a gate from discrete components.
 
Im discussing this problem on another forum in parallel with this one.
I just copied what was there into here.
---

So the big problem I am facing today is to make this 7493 IC count to 7....
In the proteus simulator, is working fine, but in reality, it stops to 4, doesnt matter the wiring I put on it !!!
I tried separately this circuit and now is counting up to 16, for 2IC's 7493 - the one I was using all the time and a brand new one.
1663666890787.png

So this is telling me the 7493 is a good chip, undamaged !!! GOOD !
Then, I switched to another NAND gate on the chip. Also I switched to another AND gate on the chip. You know, to try other gates from the 4 they have, in the idea maybe 1 gate is busted but the others are good.
I got the same result. It is counting and stops at 4.
It is counting to 3 actually (lit 2 leds), but when is trying to count to 4, is resetting. Like the wiring is connected for a 4 counts.... like this:
1663666922444.png

I became suspicious on my AND IC and I read carefully on it again, and it says CD4081BE. So it appears to be the right chip for an AND gate... But it behaves like a NOT gate...
It might be what you just said mister @dl324, being CMOS, easily damaged by ESD, either by me or from the guy I take it from, or board I scrap it.
---
I believe it is time to make a NEW and Glorious AND gate from discrete components. I only need 1 so its not a big deal.
I never built any logic gate before. This is my first time. And my first instinct is to make if from diodes. Like in this example:
I assume that RL is a 10k?
1663666947712.png

I am also thinking to test both IC's -theNAND and theAND- and see if they are doing what they should...before building a discrete gate.
And I just tested theAND gate. It does what it say it does. When A and B are to gnd, or one is on (+) and the other is on (-), the J output stays LOW. But when both A and B are to High, the output J gets High.
BUT... with an interesting side effect. When 1 or both A and B inputs are not connected to (+) or (-), when they are floating, the output J gets High....
Ta-ta-tam !!! This may be the reason mister @dl324 was saying these chips categories dont match eachother? the TTL and CMOS? May be.
One answer is certain. If one of the inputs (or both) gets in a tristate mode (not - not +) then it's output is High, and is certainly confusing/overriding the counter.
1663666968138.png

So, I solved this --floating problem-- by inserting 2x10k on each input pin (on A and B) and indeed solved it. I only had to put 1x10k on a single pin and the output wasnt High when floating. Also, it didnt affected the normal operation at all. Only when both A and B were High the output got High. The rest of the cases, is staying Low. Which is good.
I immediately inserted into my counter circuit. Absolutly no effect, it is still counting to 4. Incredible !!!
I also put 10k on the NAND inputs, and outputs, everywhere. No effect on the counting at all, still counting to 4. Again, incredible !
What a puzzle to descifre!
 
To count from 0000 up to and including 7 ie 0111 in binary the reset to state 0 ie 0000 takes place on the 8th state ie 1000. Thus connect both reset pins to Qd.
 

Attachments

  • 3C383991-6466-40FD-A1F1-69E3DA3D5CCC.jpeg
    3C383991-6466-40FD-A1F1-69E3DA3D5CCC.jpeg
    382.4 KB · Views: 48
Ok, at the moment it is counting to 8. If you number from 0 to 7 you will get 8 pulses.
I need 7 counts, that means from 0 to 6. Right now, when is reaching 6, leds stay open but in the moment is counting 1 more up, to reach 7, to light all those 3 leds, it should imediately reset to 0. As usually does when is reaching a certain programmed value. But in my case it is litting all 3 leds on 7, sign that is actually counting to 8. So right now is still not good.
Unless it is some trick that you know and I dont?
1663673699315.png
 
For seven counts from state 0 to state 6 the reset to state 0 happens when the output state is 0111 ie on the seventh state.

Thus one must AND the outputs Qc, Qb and Qa to produce a reset pulse when they are logic 1. The reset pulse goes to both R1 and R2.

To avoid timing errors we should AND these three Qs in one gate. Do you have any three input ANDs, NANDs, ORs or NORs?
 
Dispense with the first ff stage ie clk a and Qa. Feed clock into clk b. Use the three ffs of second stage ie Qb, Qc and Qd only as the 1, 2 and 4 outputs respectively. Connect reset R1 and R2 together and to a logic circuit which provides a logic 1 when b, c and d are all logic 1.
 
Last edited:

OFFICIAL SPONSORS

Electrical Goods - Electrical Tools - Brand Names Electrician Courses Green Electrical Goods PCB Way Green 2 Go Pushfit Wire Connectors Electric Underfloor Heating Electrician Courses Heating 2 Go Electrician Workwear Supplier
These Official Forum Sponsors May Provide Discounts to Regular Forum Members - If you would like to sponsor us then CLICK HERE and post a thread with who you are, and we'll send you some stats etc

Advert

YOUR Unread Posts

Daily, weekly or monthly email

Thread starter

_q12x_

DIY
Joined
Location
New York
If you're a qualified, trainee, or retired electrician - Which country is it that your work will be / is / was aimed at?
United States of America
What type of forum member are you?
DIY or Homeowner (Perhaps seeking pro advice, or an electrician)

Thread Information

Title
AT28C16 EEPROM learning experiments
Prefix
N/A
Forum
UK Electrical Forum
Start date
Last reply date
Replies
121
Unsolved
--

Advert

Thread statistics

Created
_q12x_,
Last reply from
_q12x_,
Replies
121
Views
10,349

Advert

Back
Top