User talk:Thingmaker

Welcome to Wikipedia again!

"I always liked the impossible... I found quite often impossible problems could be solved with extremely simple methods..."

How wonderful that sounds, William! Wikipedia should be proud when such remarkable personalities as you want to participate in it!

A few years ago, I followed with interest your professional comments, especially on the talk pages dedicated to TTL and ECL logic gates (e.g., in Archive_1). Unfortunately, no one paid any serious attention and do not reflect them in the Wikipedia articles...

I (Circuit dreamer) contributed closely in Wikipedia during 2006-2010 and then I managed to make a lot in its circuit section; see for example my list of contributions... and especially the articles about digital circuits.

But my efforts to find clear, simple and powerful explanations of basic circuit solutions were perceived by Wikipedians as original research. As a result, I was banned from editing and actually I abandoned Wikipedia...

Then I (Circuit-fantasist) entirely moved to Wikibooks where I created Circuit Idea - a book about great circuit ideas; see for example the story about ECL.

Best regards, Cyril (Circuit dreamer).

Thanks Circuit dreamer. I guess there are two of us banned from Wiki's logic circuit descriptions. It appears they prefer information from amateur handbooks to first hand knowledge. I just made a new comment about the DTL description. It is a disaster! I will check the sits you referenced. ThingMaker

I just looked at your Archive_1). It makes me want to go back into hiding!ThingMaker

Diode-transistor logic

Just to let you know that I moved your comment from the article itself on to the article's talk page : Noyster (talk), 22:02, 21 July 2014 (UTC)[reply]

I don't understand! As far as I know I put my comments on the Talk page. If I placed it on the article I regret the error.ThingMaker

Hi Thingmaker. Your story about the development of logic gates is just incredible... you have to publish it in some more suitable and reputable place... I just want to say a few words in defense of what is written by me in the DTL page:

"The two resistors R3 and R4 form a resistive summing circuit with weighted inputs that adds the negative bias voltage V− to the positive diode logic output voltage... As a result, the unipolar (positive) diode output voltage (about V+ for logical one and 1.0 V for logical zero) is converted into a bipolar voltage (a few volts above and below ground) to drive the output transistor"

I understand your reaction since the written is a bit unconventional explanation... but to be honest, it is still true... You are right that "R3 and R4 provide a current sum to drive the base"... but it is also true that actually they add voltages... R3 and R4 act as voltage-to-current converters that convert the diode output voltage and the negative bias voltage into currents... then these currents are summed (subtracted)... and the current sum drives the base. This summing circuit is so important that I have dedicated a separate story about it in Circuit Idea wikibook.

I understand very well the purpose of this resistive network and the additional negative supply - to drop the base voltage below the zero voltage... to make it negative... and thus to cut-off surely the transistor (especially if it was a germanium type)... You can think of this circuit also as a voltage divider with a "shifted ground". When its output voltage exceeds V_BE0 (appr. 0.7 V), it is shorted by the forward-biased base-emitter junction; when it is below this threshold, the base behaves as an open circuit. A similar but not so reliable effect can be achieved by replacing R3 with a diode or a base-emitter junction (as in TTL).

I have considered this bias technique in RTL article as well. Regards, Circuit dreamer (talk, contribs, email) 07:28, 22 July 2014 (UTC)[reply]

Please indent your replies on talk pages

See WP:INDENT. SpinningSpark 13:12, 10 August 2014 (UTC)[reply]

Should I clean out my sandbox and go home?

As I said on the Diode logic talk page I am wondering if I should just clean out my sandbox and go home. I have a great deal of experience and knowledge in electronics and some related fields that I believe Wikipedia could benefit from. I believe I qualify as an expert on discrete transistor circuit design. There are few of us left. I had a very unique and successful career with IBM ranging from transistor circuit design through fixing a difficult problem with TI’s integrated circuit design (while working for IBM) on to inventing the capacitive keyboard, defining a satellite communications system and digital signal processing including the speech processing algorithms. I trained and taught the circuit design engineers that supported the IBM Research Triangle Park Lab and even years later was considered the expert that often solved their impossible circuit problems. After my age of fourteen I learned relatively little electronics from class rooms and even less from books. On a few occasions an engineer might tell me something he learned in school then I would improve on that to make a better MODEM or speech coder. I have forty-five “formal” inventions, thirty of which resulted in patents and fifteen in patent publications. Nearly all in the field of electronics with many in electronic circuit design.

I only tell you this to establish my strange background. My knowledge was sufficient to carry me to a very successful career. I know a lot of things. I can’t tell you where I learned them because I seemed to just know them. That knowledge has been tested in the battle to succeed when competing with thousands of electrical engineers with conventional educations.

Much of the confusion with the Wikipedia transistor logic circuits is due to the fact that several of the types (RTL CML and DTL) came from the world of discrete transistor circuit design. They are misunderstood in the world of integrated circuits. Many of the published citations came from books written for hobbyists. Some come from component handbooks written to encourage the use of their products with no knowledge of a practical design. Both often present circuits that won’t work and were never used in a serious application. I know these circuit families from a practical application view point and can fix the articles. Some of these circuits can be properly cited like CML. Some were kept trade secrets because of the competitive nature of the business during the discrete design era.

I am not a research journalist. I am not skilled at searching literature for information I already know.

For the most part I hope to offer information that is recognized and accepted by those skilled in the art. It doesn’t seem like such information needs citations especially irrelevant ones to make them appear authentic.

So I ask, does Wikipedia want my help?Thingmaker (talk) 14:20, 23 August 2014 (UTC)[reply]

A miracle in Diode logic

Thingmaker, you made a real miracle in Diode logic talk page - you made electronics Wikipedians begin reasoning with their own mind... and not just blindly quote someone else's phrases... I tried to do it for five years... but I did not succeed... Circuit dreamer (talk, contribs, email) 12:30, 24 August 2014 (UTC)[reply]

Thanks! I am trying!Thingmaker (talk) 14:30, 24 August 2014 (UTC)[reply]

I am also trying to reason about this elegant simplicity - diode logic gates:)

About the extended diode biasing (+12V instead +6V and -6V instead 0V)... It seems it solves sooner the problems introduced by the relatively high pull-up and pull-down resistors than by the non-ideal diodes. But real diodes can cause some interesting effects. Thus, if we cascade a few diode ANDs, the logical 1 voltage will continuously increase above the nominal +6V voltage (+6.7V -> +7.4V -> 8.1V... for Si diodes); if we cascade a few diode ORs, the logical 0 voltage will continuously decrease below the nominal 0V (-0.7V -> -1.4V -> -2.1V...) It seems we should (if possible) alternate AND and OR gates...

Current steering... In your extended bias configurations, at least one diode is open - in AND gate, this is the diode controlled by the lowest input voltage; in OR gate - the diode controlled by the higher input voltage. So, the AND circuit always "pushes" a current into while the OR circuit always "sucks" a current from some (or all) input(s)... Even when all the input voltages are high (in AND) or low (in OR) the current has to pass from somewhere... and if there is at least small difference between the input voltages, it steers between the input sources... Circuit dreamer (talk, contribs, email) 14:59, 24 August 2014 (UTC)[reply]

Don't think of a diode as a switch. It is just a nonlinear resistance that conducts current in the forward direction with very low voltage drop and in the reverse allows very little current at any voltage (ignoring breakdown). The diode with the most positive input for the OR or most negative for the AND simply pulls the output with it. The current changes or switches, if you prefer, from one diode to another but the diode does not act as a switch. As far at the accumulated diode drops for several stages I have never seen more than two stages, AND and OR or OR and AND used at one time with simple diode logic because the signal gets too degraded. Also without special design the AND can not drive the OR or vice versa because they don't have enough output current drive. The +12 and -6 volts help this.Thingmaker (talk) 20:02, 24 August 2014 (UTC)[reply]

Thingmaker,

What do you have against this to think of the diode as of a switch? "Diode" is a general, neutral, meaningless name of this 2-terminal element while the generic name "switch" shows its function in this application - to commutate the current. We can use a diode as a switch (if we neglect V_F)... or as a voltage stabilizer (if we pay attention to V_F)... or as a log/antilog converter (if we consider the exponential shape of its IV curve)... or as something else... IMO here the diode acts exactly as a switch, not as a voltage stabilizer or a non-linear element...

There is an interesting contradiction here. If an AND gate drives an OR gate, the pull-down resistor should have many times higher resistance than the pull-up resistor not to decrease the logical "1" voltage and v.v., if an OR gate drives an AND gate, the pull-down resistor should have many times lower resistance than the pull-up resistor not to increase the logical "0" voltage. Regards, Circuit dreamer (talk, contribs, email) 20:55, 24 August 2014 (UTC)[reply]