Talk:Diode logic

The mystery of diode AND gate


Diode logic gates were introduced in 50's; I met them for the first time in the late 60's when I went to technical school. I remember I understood the explicit diode OR gate but I was in a maze by this so odd, exotic and absurd diode AND gate... I asked myself many questions as: "Why the diodes were back to front? Why the resistor was connected to +V instead to ground? Why there was no input current when the input voltage was high? And why there was input current when the input voltage was low? But why the current went out of the diodes and went in the input source? What was this absurd? Why it was impossible to make inverting diode gate? Why AND gate was supplied by an additional voltage source while OR gate had not such a source?...and so on...and so on...

It is interesting fact that many years later I continued to not understand what the basic idea behind diode AND was. I perfectly knew the circuit but I didn't understand it... I "saw the trees but I didn't see the forest for them"... And what is more interesting I didn't see any reasonable explanations of this humble circuit somewhere to answer my childhood questions... instead I was seeing thousands of "not seeing the forest" explanations...three months ago), during the laboratory exercises with my students on Digital circuits, I began realizing the great idea behind these legendary diode circuits. I was amazed how simple it was. I shared and considered my insights with these young people and they approved of them; they admired this "elegant simplicity". We decided to share the truth about diode AND gates with wikipedians and Wikipedia readers. That is why, I have written these explanations in the article. In addition, I have listed (in bold) and explained (in italic) below the key points of understanding diode AND gates extracted from the main article.

Is it an original research? I have already explained it in the article about Miller theorem. Yes, it is... but in the common sense of this term, not in the specific Wikipedia OR sense. Really, I have exposed my insights about this circuit solution but I have managed to reduce them to extremely clear, evident and simple explanations that should not be treated as original research in the Wikipedia OR sense; they do not need to be referenced. If you do not agree with me, please insert your comments below the items. Circuit dreamer (talk, contribs, email) 18:10, 15 January 2011 (UTC)

Key points of explaining diode AND gate

 * In logic gates, logical functions are performed by parallel or series connected switches controlled by input logical variables. ''It is well-known truth, see also, , , , etc.


 * In diode logic, the switches are implemented by diodes: when forward biased, a diode switch is closed; when backward biased, the switch is open. ''It is well-known truth, see also, , , etc.


 * OR logic gates are implemented by parallel connected switches. It is well-known truth, see the sources above.


 * In a diode OR logic gate, diode switches are connected so that they act as normally open switches: if the input voltage is high (input logical 1), the according diode switch is closed; if the input voltage is low (input logical 0), the diode switch is open. It is obvious, there is no problem to understand it in the case of a diode OR gate.


 * Diode OR logic gates consume current from input sources with high voltages and they do not inject current to input sources with low voltages; so, they belong to so-called current-sourcing logic. We will discuss this phenomenon thoroughly below; for now see . See also the text I have extracted from this forum (we will use it when explaining diode AND gate):


 * "...If you take one of the simple power source, a battery. If the battery is 9V, and you put a load across the two poles, then the current flows through the load. The current will go in one direction. One pole could be referenced as GND, the other as Vcc. Now, take two batteries - one is the above 9V battery, the other is a 12V battery. You connect the negative poles together, let it be the GND reference. Then, if you connect a load between the two positive poles (one pin on 9V battery + pole, the other pin on the 12V battery + pole), then, current will flow, but will go in the *other* direction, in the 9V battery, than in the first case. When the 9V battery provides current, which goes out of 9V (using conventional current), the battery is said to *source* current. When the battery is absorbing current, it is said to *sink* current. Source and Sink really only does tell in which direction the conventional current flow..." (I have corrected only some spelling mistakes)


 * AND logic gates should be implemented by series connected switching elements. It is a well-known truth, see the sources above.


 * In contrast to transistors, diodes are two-terminal switching elements, in which the input and output are not separated; they are the same. As a result, series connected diode switches cannot be driven by grounded input voltage sources. The 4-terminal relays are the most suitable electrically-controlled switches since their input and output circuits are completely separated. The 3-terminal transistors can still be used as 4-terminal switches by grounding the emitter (source); thus, the input and output circuits have a common point (the ground) but they do not interfere in each other's operation. The situation with 2-terminal diodes is the worst - here the input source and the load are connected in the same circuit and there is no way to connect the diodes in series and to drive them by grounded voltage sources.


 * To solve this problem, diode AND logic gates are constructed in the same manner as OR diode gates - by parallel connected diode switches (the same idea is used in the input stage of TTL logic gates where base-emitter junctions are connected in parallel). If you find it difficult to see the parallel connection, imagine that all the input voltages are low (i.e., all the inputs are connected to ground).


 * In a diode AND logic gate, diode switches are connected so that they act as normally closed switches in respect to the input voltage: if the input voltage is high, the switch is open; if the input voltage is low, the switch is closed... To realize this idea, the diodes are reverse connected and forward biased by an additional voltage source +V (a power supply) through the pull-up resistor R1. This is a key concept in understanding the diode AND logic gate; so, I will explain it thoroughly.


 * The name of this circuit technique is biasing (it is not only electrical; it is a general idea that can be seen around us in this world). To be more precise, this is "biasing with the same but opposite to the input quantity" with the purpose to reverse the behavior of some element. In this arrangement, we apply not the genuine input voltage across the element; we apply the difference between the input and biasing (offset) voltage, i.e., we apply the inverted input voltage. As a result, the current flowing through the element is inverted as well.


 * ''First, see the extracted forum text above that explains the same concept. Or imagine another problem: we have a relay with normally open contact but we need a relay with normally closed contact. What do we do? We can invert the relay behavior by connecting the one coil terminal to +V instead to ground. This means that we have inserted an additional voltage that is equal but opposite to the input voltage (traveling along the loop +V -> VIN -> coil). As a result of this biasing, when applying +V to the other coil terminal, the resulting voltage drop across the coil will be zero and no current will flow trough it; and v.v., when applying 0 to the coil terminal, the resulting voltage drop across the coil will be -V and current will flow trough it in opposite direction (+V -> coil -> input source -> ground). The relay with normally open contact has become a relay with normally closed contact.


 * We may see the same trick in a CMOS logic gate, in a PNP transistor stage or in any complementary transistor stage that are driven and supplied by positive voltages. In these circuits, the upper PMOS or PNP transistor is actually driven by negative input voltage (with respect to +V), which is the inverted positive input voltage. From this viewpoint, a diode AND gate with positive input voltages is actually a diode OR gate with negative input voltages.


 * '''To obtain AND instead OR function according to De Morgan's laws, the input and output logical variables are inverted:


 * Y = NOT ((NOT (X1) OR NOT (X2)) = NOT (NOT (X1 AND X2)) = X1 AND X2.


 * Finally, we have only to describe the diode AND structure by logical expressions and to simplify them by applying De Morgan's laws. There are a lot of sources about such transformations (e.g., ).


 * Therefore, the diode AND logic gate is a modified diode OR logic gate: the diode AND gate is actually a diode OR gate with inverted inputs and output. This final conclusion evolves from the considerations above.


 * As diode AND logic gates do not consume current from input sources with high voltages and they inject current to input sources with low voltages, they belong to so-called current-sinking logic. We have just explained this phenomenon above. There are also a lot of sources about current-sinking logic.

Circuit dreamer (talk, contribs, email) 18:10, 15 January 2011 (UTC)

The main prerequisite for editing in electronics area
Wtshymanski, I know how to answer in detail to your provocative action but I will not do it since persons as you deserve no such attention; it would be dead loss of time for me. I will only remember to you that the first prerequisite to do something different than cosmetic edit here, in electronics area of Wikipedia, is to have some elementary notion about circuit ideas. But you obviously have no any idea about circuits since if there was even the smallest glimmer of circuit thought in your mind, you would comment the comprehensive explanations above written especially for you and your likes. Circuit dreamer (talk, contribs, email) 11:59, 16 January 2011 (UTC)
 * You compress less meaning into more words than most other editors I've seen. Wikipedia can always buy another terabyte drive for a few dollars, but the reader's time is scarce and valuable, and your long rambling discursions add no explicative power to the article and waste the reader's time. --Wtshymanski (talk) 14:42, 16 January 2011 (UTC)


 * The difference between the two approaches is that you and the most Wikipedia editors describe WHAT is made while I explain WHY and HOW it is made. For example, if you describe the diode AND gate you will ascertain THAT the diodes are turned with the cathodes to the input sources and the resistor is connected to +V instead to ground while I have revealed WHY these changes are made. Another difference is that I always generalize circuits, elements and the ideas behind them. For example, you will probably say "diodes" while I have said "diode switches". This allows to make connection with other implementations of the same circuit idea where other elements (e.g., relays, neon lamps, BJ and MOS transistors, etc.) act as electrically controlled switches.


 * As a result, in the first case readers will KNOW circuits while in the second case they will UNDERSTAND them. But obviously, the truth is somewhere in the middle. So, I appreciate your ability to write clearly and concisely and I also make efforts to improve my style along these lines. Circuit dreamer (talk, contribs, email) 22:40, 17 January 2011 (UTC)

Question and comments
I have a question and a few comments.

1. When was DL invented? Is there a specific date?

2. "output logical 1" and similar expressions in the descriptions of OR and AND operations are repetitive. Stating what 0 and 1 mean once is sufficient.

3. "Powerful diode OR circuits are used in the simple UPS to switch between the main and the standby supply sources." I believe it should sound more like "Powerful diode OR circuits are used in simple UPSs to switch between the mains and the standby supply sources."

ICE77 (talk) 22:01, 16 February 2011 (UTC)

Mickey Mouse logic
With respect to the name "Mickey Mouse logic", whose addition reverted, the person from whom I first heard the term told me it came from some CMOS cookbook or TTL cookbook by Don Lancaster. --Damian Yerrick (talk | stalk) 16:53, 26 September 2011 (UTC)
 * The problem is, that's about the only place you find that term used (or, people quoting Lancaster). It's not a widespread phrase. --Wtshymanski (talk) 17:49, 26 September 2011 (UTC)

Diode Logic, a More Conventional Explanation
I apologize for the length of this addition to the main article. When I first saw the article on Diode Logic I felt it needed a simpler, more conventional explanation. The original article seemed a bit more "philosophical" than a young mind might find useful. I was encouraged (maybe that is too strong) to write it. About two paragraphs should do it. Then I considered who might read it. I found my way here out of curiosity, what could be said about such a simple circuit. On August 5, 1957 just out of the army and as an electronics technician never having heard of logic circuits I was told to redraw some block diagrams for a design. When I saw blocks labeled AI, OI, -AI and –OI I asked what they were. The engineer sketched a rough schematic of a Diode AND Invert circuit. Its operation was immediately obvious. If the NAND circuit is so simple who would be interested in this article. I concluded the most interested might be someone with no knowledge of logic circuits and possibly very little knowledge of basic electronics. This is not the place to start from basic electronics but I hope by at least explaining the diode someone with very little electronics knowledge might understand this basic function. I hope I achieved my goal. If my contribution is determined to be less than appropriate it can always be deleted. I hope it isn’t. I wish I could have read it when I was eleven years old and in sixth grade. Incidentally one year after that first introduction to the NAND Gate I started designing whole logic families of them. I would still be a technician for another year but I was doing a Junior Engineer's job. Thingmaker (talk) 19:26, 12 August 2014 (UTC)


 * It appears someone removed my contribution to Diode Logic. Can anyone tell me why?Thingmaker (talk) 21:09, 13 August 2014 (UTC)


 * I was brought here because someone left a note at the eelctronics wikiproject. Did you read the edit summary from the person reverting you?  Wtshymanski can be a combative editor, but in this case I concur, creating a fork of the same content is not helpful and against guidelines.  No doubt there are a lot of improvements that can be made here, but they should be made incrementally to the existing material.  I have a few other comments on your text
 * The whole approach seems to be from the point of view of trying to teach the subject. Wikipedia is not a textbook.
 * The detailed description of the diode should be drastically summarised and the reader linked to the diode article for further details.
 * You need to read MOS:CAP
 * Hope that helps, SpinningSpark 00:03, 15 August 2014 (UTC)

DL or DRL
This message is primarily for User:Spinningspark but other opinions are welcome.

For the most part I do not feel I am an expert on naming conventions and to a reasonable extent that applies to naming logic circuit families. I spent three years primarily designing CML, DTL, and RTL circuits and related functions. I spent my first eleven years designing transistor circuits of some form or another. Texas Instruments TTL circuits were the first time I heard any of the above naming conventions. Because of this I proclaim not to be an expert and will yield to your judgment.

That said I offer these thoughts: If DL is DRL then shouldn’t DTL be DRTL and TTL be TRTL and maybe RTL be RRTL. Should a simple inverter be RTL since it has resistors and a transistor? Would this make the other RTL less meaningful? It was always my impression that the first letter, D, T, R signified the component that performed the logic. The final T indicated the type of amplifier or inverter. I always assumed the R’s used for biasing were omitted because they were omnipresent in nearly all bi-polar transistor circuits and nearly all other circuits except CMOS type logic. In support of your definition is the transistor logic used in the Apollo Guidance Computer. That circuit has been called RTL which has always bothered me. It is in fact simply an inverter Dot OR logic where the logic takes place by multiple transistors sharing one load resistor. I would call that TL or possibly TTL  (I am not trying to teach you just offering thoughts.)

Could there be a standard that explains and defines the rules?Thingmaker (talk) 14:22, 19 August 2014 (UTC)


 * For those who might be confused by Thingmaker's post above, I guess it is in response to my restoration of the term "diode-resistor logic" in the lede. Thingmaker, this is not a dispute over the title of the article.  I am perfectly happy with it being called diode logic.  This is simply a matter of house style.  The convention is to bold the title term of the article in the lede sentence.  We also bold alternative terms that redirect to the article.  Diode-resistor logic is such a redirected term.


 * The governing policy on names is Article titles and its subsection WP:COMMONNAME is pertinent here. We use the common name found in reliable sources.  Thus, we would not use DRTL or TRTL since these are not, as far as I know, commonly used in book sources.  However, DRL is found in numerous textbooks so is at least a valid redirect, and arguably could be the name of the article (although, as I said, I am not proposing a change).  All the books I have looked at define DRL as the diode OR and AND gates described in our article.  I am not sure whether or not you are arguing that there is a substantive difference between DL and DRL.  Even if there is a difference DRL should still be bolded in the lede along with the defining difference (which I would want to see referenced because I am not convinced there is a difference). SpinningSpark 15:07, 19 August 2014 (UTC)


 * I see your point,  I tried searching for diode-resistor logic in Wikipedia and it took me to Diode logic.  Does the “From Wikipedia, the free encyclopedia (Redirected from Diode-resistor logic)” at the top of the page automatically create the re-direct page that automatically took me to Diode logic?


 * I would still like to ask the questions about naming logic families. I keep seeing someone place the R in DTL and calling the Dot OR transistor logic RTL.  I wish someone could find an official answer.  I guess since DL or DRL might fall into a different category I shouldn’t be too bothered with that.


 * My question for the "Talk" audience is, “What is the convention for naming transistor logic circuits?” Does the first letter signify the primary logic device (D, R, T, whatever) and the second signify the amplifier device?  L of course signifies that it is a logic circuit.Thingmaker (talk) 19:30, 19 August 2014 (UTC)


 * For how redirects work see Redirect. We don't have, as far as I know, a guideline on naming logic families so WP:COMMONNAME is the guiding principle.  I don't think you can characterise these abbreviations as "logic device/amplifying device".  DRL, or for that matter RRL, does not have an amplifying device at all.  I think it may be more on the lines of "input device/output device" but there is really no guarantee of consistency here anyway. SpinningSpark 20:04, 19 August 2014 (UTC)

Note
It should be noted that this article has been significantly changed as of August 20, 2014. The purpose is to correct the theory that a diode is a switch to the p-n diode theory that a diode is a passive nonlinear impedance with a low impedance and low voltage in the forward direction and a very high impedance in the reverse.Thingmaker (talk) 14:10, 20 August 2014 (UTC)
 * It's OK but not mandatory to describe changes you've done; the edit history is there for all to check. I think the reason that textbooks use "switch" is as a shorthand for "nonlinear passive...etc." - which is mouthful to say and is just one way of thinking about a switch anyway. --Wtshymanski (talk) 16:16, 20 August 2014 (UTC)

World's largest computer
I have removed the following passage from the page:


 * In 1957 one circuit generating the clock for the “World’s largest transistorized computer” was used to pass a half microsecond pulse. It failed after three years but fortunately faster transistors were available to fix it.

I intitially marked this with cn and on the claim in quotes cite quote but both templates were removed without providing the necessary citations. I think the passage is problematic so am bringing it here for discussion. I realise that the article is almost entirely devoid of references but it can largely be found in numerous textbooks so I am not so concerned with the general lack of references. On the other hand, anecdotes of this sort definitely need sourcing. The rest of the section remaining in the article really needs sourcing as well.

The claim to be the "world's largest transistorized computer" in particular needs a source. I also asked for the quote to be attributed. However, the latest version would seem to indicate that this is not a quote but is WP:SCAREQUOTES and it is not really the largest. Starting off by saying this was the first transistorized computer kind of makes it meaningless anyhow. It would also have been the world's smallest transistorized computer. It is really not appropriate to have this in the article at all as far as I can see. It is also inappropriate to cite this to "the author of this article". Firstly, because anecdote is not a reliable source, but also because it is WP:OWN and self-referential. SpinningSpark 22:52, 21 August 2014 (UTC)


 * 1. I would have no problem with removing the whole subject “Early diode logic with transistor inverter”.  I had some doubts about adding it but I felt it would add some meaning of what diode logic might be used for.


 * The “world’s largest computer” in August 1958 was not the IBM 608. It was a special purpose product developed under the leadership of Bob O. Evans in his “Navy Department” at the IBM Glendale Lab in Endicott NY for a secret US government organization (not the navy).  It used 20,000 transistors.  The 608 used 3,000 transistors.  Maybe if the claim of world’s largest was removed it would be more acceptable.  Or just delete whatever you think.  Incidentally, the claim of the 608 being the first all transistorized came from the Wikipedia article and also the IBM 608 site.  I have no personal knowledge of that.


 * 2. I removed the “citation” notices because I thought I provided them. I doubt if there is any public record of this machine since it was secret.  I was one of about fifty people involved in the project.  I stated my reference was a personal firsthand account.  Ignoring whether my attempt for a citation was acceptable, should I have just added it and left the citation request.  I thought if I removed it you would be alerted and would do as you did.


 * 3. My diode logic article has few citations. It consists of common knowledge that I would think would not need many citations.  It seems common practice to place citations that only seem to prove a point or point to an article that is too large to search for the reference.  I used the p-n diode citation as you suggested.  In fact it has very little of use.  It does a fine job of presenting the solid state physics of a p-n junction but little that directly relates to an external understanding of the diode.  I am hoping to add a graphic interpretation of Shockley diode equation that is more directly usable for a circuit designer or technician.  The article has only one plot of a diode curve and that is grossly distorted.


 * If two people that understands the electronics of diode logic believe the previous version was more accurate than my work then it should be removed all together.Thingmaker (talk) 18:03, 22 August 2014 (UTC)


 * There's no need to talk about deleting your work in its entirety, I'm not suggesting that and I encourage you to carry on editing. But I think you need to read Reliable sources.  Wikipedia articles are never acceptable as references, otherwise we could end up with Wikipedia articles entirely self-referencing themselves in one big circle.  Personal recollection is never acceptable as a reference; all references need to be verifiable, a core policy here.  I agree with you that there is a lot of unnecessary citations on Wikipedia, but if you ever put up an article for review, such as for Featured Article status you will find that the expected standard is a citation for every sentence.  That is not what our policy requires, it is much less onerous than that.  Personally, as I said above, I am happy to leave uncited anything that can easily be found in standard textbooks, but if you want to get an article past review you just have to suck it up on that. What is unacceptable and will eventually get deleted by someone or other is material that it is not possible to cite. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 21:43, 22 August 2014 (UTC)


 * I see you have now deleted the whole section. I think you give up too easily.  This search on gscholar, and also this from gbooks, indicate that something might be found in sources.  Most of it is behind paywalls, but the librarians hang out at WP:REX and can help out with that if you want to do some research. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 23:52, 22 August 2014 (UTC)


 * As you have seen I restored the “Early logic” portion and changed some wording hopefully to make it less controversial. I added some citations for whatever they are worth.


 * I am quite certain there is no public information that references the 20,000 transistor system I mention. An hour after I installed my faster circuit I had a loaded M1 Carbine rifle pointed at me.  An hour and a half later the duty officer determined I didn’t need to be shot because a marine had issued me the wrong badge requiring that I have an armed escort.  This was a highly secure government secrete facility.  I saw a vague reference to IBM pursuing a Navy contract.  The machine in question was developed in Bob Evans’ “Navy Department” but the project did not involve the navy.


 * I have no problem with you or whoever changing or deleting what I presented in this section. I only added it for its possible interest value.


 * I hope to add, on your “Talk page”, a discussion about what I have to offer Wikipedia. Possibly I should clean out my sandbox and go home.  I hope not!  The electronics sections need some help especially in the logic circuit area and I have the knowledge to fix them if the rules allow it.  The rules seem to me more ridged in the electronics area.Thingmaker (talk) 13:02, 23 August 2014 (UTC)

Ideal diodes? And logic levels?
"But why do we need to pull down to -6 volts?" says our hypothetical bright 12-year-old reader, "Doesn't that make the output of the gate -6 V for logic 0, not 0 V ? " And if we have ideal diodes anyway, we don't need the pull-down to a negative voltage. Once we explain the topology of the ideal gates, then we can get to the ugly reality of non-ideal devices with finite switching speed and leakage current, capacitance, minority carrier storage, resistance (of the lossy variety, forsooth) and so forth. A description of how these non-idealities militate against use of diode logic in complex systems would be quite encyclopedia worthy. --Wtshymanski (talk) 02:11, 24 August 2014 (UTC)


 * Are you asking a hypothetical question (on behalf of your 12-year-old) or do you really not understand? I suspect that you do understand, but anyway, the logic 0 output is not -6 V, it is 0 V (less one diode forward drop), assuming that the gate input is a true 0 V.  Even with ideal diodes, the pull down is still doing something.  Without it, both sides of the diode are connected to 0 V, the diode is sitting right on the origin of its I-V curve, and it is thus undefined whether it is turned on or not.  It is thus undefined whether the output is connected to the low impedance source (via the diode) or through the load resistor to the 0 V rail.  The pull down guarnatees low impedance.  I think it is good to have real circuits in the article.  Having said all that, I agree with you that basic circuits should be presented first, free of all extraneous items like output transistors.  If for no other reason than book sources invariably show the simple circuits in dealing with this subject.  Otherwise, this page is in danger of covering the same ground as diode-transistor logic. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 08:10, 24 August 2014 (UTC)


 * I questioned the need for DL myself. I see no problem with it existing but it makes more sense when presented with a practical application such as DTL.  By itself it has very limited application.  Thanks Sparks for providing your response.  Adding more to the answer for the 12 year old, if RD returns to ground it has no output drive at the 0 level which would make it even more useless.  The 0 input allows the input diodes to limit the output to 0.  If the resistors return to 0 volts for the OR and +6 volts for the AND then neither could drive the other since they would have no output current drive at those levels.Thingmaker (talk) 11:12, 24 August 2014 (UTC)
 * If it's not in the article, it ought to be. Why -6 volts? This is the problem we get into when we start mixing our "ideal" diodes with real world circuits. Since we're assuming ideal diodes with zero resistance and no voltage drop, there seems to be no motivation for the pull-down; we don't need "output drive" since that's only necessary for non-ideal diodes. But as the reader can see from the 608 gate schematics, real diodes need a pull down. --Wtshymanski (talk) 13:56, 24 August 2014 (UTC)
 * As I said above, the pull down still makes sense with ideal diodes. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 14:50, 24 August 2014 (UTC)


 * Wtshymanski are you suggesting we should have a section that discusses the non ideal diode and its implications to diode logic design. I am working on a section I hope to add to the p-n diode article that addresses the p-n diode from a circuit design standpoint (connecting the equations to real life).  I could probably write a condensed version for the diode logic article or wait and simple provide a link to the final version, it you allow it to stand.Thingmaker (talk) 14:44, 25 August 2014 (UTC)


 * Ideal diodes still must work in a real world of electronics. They have a zero forward voltage and zero reverse current. They still need a resistor to keep the output from drifting away and it still needs output drive if it has any value except in a lab experiment which would still need an output resistor.  I believe even with an ideal diode the designer would choose voltages and resistors.  I have corrected some of the changes to reflect this fact.Thingmaker (talk) 21:59, 25 August 2014 (UTC)

RD vs RD
The text for RD had been changed to RD. This refers to the RD in the schematics. I know of no convention that tells a circuit designer how to name his components other than the obvious such as R, C, L, Q and T. When I created the schematics for the AND and OR circuits I initially used Rd but found that the d was not legible when the image was shrunk. A subscript would be even less legible. Because of this I chose RD. I plan to revert to RD. If this is in violation of a Wikipedia convention then I will change and upload new schematics probably using simply R. This will leave clutter in Wiki Commons but I guess that would be better than using a wrong convention. Whoever says it should be RD can change the text or tell me to and I will fix the schematics. To keep it sensible they should both be changed at the same time.Thingmaker (talk) 19:22, 24 August 2014 (UTC)


 * The suffix is going to look fine on the diagrams. There is no need to create clutter on Commons, a new version can be uploaded to the same file name.  Ask if you can't figure out how to do it.  RD is just dumb if you want clarity, the universal convention is to use suffixes.  Regarding "Whoever says it should be RD", you can see who changed things by clicking the history tab of the article. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 21:30, 24 August 2014 (UTC)


 * I will try! Before when I tried I succeeded when it was soon after the first upload.  Another time it wouldn't allow it because it was in use.  I will try.  If it doesn't allow it I will just  upload a new one.  I will simply call the resistor R.
 * It failed because you tried to do a new upload to the same name. You need instead a new version.  Click the image, click the Commons icon bottom right, that should take you to the file page, scroll down to "file history", at the bottom of that section click "upload a new version of this file", then follow the instructions on the form. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 23:19, 24 August 2014 (UTC)


 * OK. I give up.  I can upload new versions but I can't use them.  It will  say the new version is current but it still uses the old one.  I have made a mess of new uploads that don't work.  If someone knows  how to delete the new uploads I think  that would be a good start at fixing the mess.  I think I need to just upload the new files as if they were totally new.  I will try again tomorrow.  Thanks for the help and I will take any new help you can give me.Thingmaker (talk) 01:14, 25 August 2014 (UTC)


 * It often takes the servers some time to catch up with a modification to an image. I have reverted the images on Commons to what I think are the ones you wanted (please check) and purged the page so the new ones show. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 08:45, 25 August 2014 (UTC)


 * Thanks for cleaning up my mess. I didn't just sit there doing the same dumb thing.  I tried about three different paths to do the uploads.  At Times I could get the updated one to display but it was huge and if I tried to scale it it would disappear.  Is there any way to delete all those redundant copies?  I will fix the RDd's in the text.  Thanks again.
 * Commons might be willing to speedy delete the redundant files as duplicates, see Commons:Deletion requests/Speedy deletion for the process. Strictly speaking they don't meet the criteria as they are not exact duplicates.  If Commons won't speedy them you will need to list them at Commons:Deletion requests for discussion.  They definitely won't delete the various versions in a file version history, it is technically possible but they do not meet any Commons deletion policy criteria. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 11:23, 25 August 2014 (UTC)


 * I think I will wait for them to delete the redundant ones unless someone tells me defferent.Thingmaker (talk) 14:47, 25 August 2014 (UTC)
 * Unless you ask for them to be deleted it is highly unlikely that they will be. Wikipedia deletes images because they are not used in any article, but Commons does not do that, they have a different mission. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 14:53, 25 August 2014 (UTC)


 * I guess I don't care if they don't. Thanks for your help, again!Thingmaker (talk) 17:35, 25 August 2014 (UTC)

NOT AND is not NAND
A negative logic AND is not a NOT AND - it's an OR (with negative logic levels). The logic function produced by a set of diodes with common anodes depends on how you assign logic 1 and 0 to voltage levels. Getting a NOT AND without changing the representations of 1 and 0 requires an inverter, which we don't have in pure diode logic. I suppose there could well be circuits where we use positive logic definitions for the first column of diodes, switch to negative logic definitions for the second column, and so on...but I don't think I'd care to present an example. --Wtshymanski (talk) 20:50, 25 August 2014 (UTC)


 * NAND is not a negative AND it is a Negated AND or an AND Invert. I probably misused the NOT. I think I was trying to preserve the "NOT NOT" logic from the previous article.  I believe NOT is reserved for the naming of inverted signal names such as A and B going into a  NAND gate results in an output of NOT (A AND B) or NOT A OR NOT B.  An AND is a negative OR and an OR is a negative AND.  A logic designer does not need an inverter to choose to use negative logic.  He can name and change the logic levels to suite his needs.  Today with logic families providing every logic function anyone would need it is not so important.  There was a whole generation of logic designers that did their job using only AND Inverts or only OR Inverts. It was important that the AND Invert was also a negative OR Invert or they would not have had an OR function.  They just called it what they needed.Thingmaker (talk) 23:27, 25 August 2014 (UTC)
 * This has to be clear in the article; changing the representation of logic 1 and 0, for the exact same circuit, changes it from AND to OR. Got to be careful to explain that the difference is only a change in the point of view of the designer for that stage. --Wtshymanski (talk) 00:55, 26 August 2014 (UTC)

Switching speed
It would be nice to say something about the speed of diode circuits, and the limitations on speed. And I've just seen a snipped of something on Google Books that describes using tunnel diodes for logic, but it wasn't clear if it was the same sort of topology as we're looking at here...sounded quite fast for its time period, though. --Wtshymanski (talk) 01:02, 26 August 2014 (UTC)


 * I will offer a response for diode speed and tunnel diode logic soon.Thingmaker (talk) 10:48, 26 August 2014 (UTC)
 * I lived through the excitement of the Esaki diode. I believe it was spring of 1960 everyone rushed to the electronics conference in Philadelphia to hear Dr. Esaki.  He stood on the stage for the better part of a hour not able to speak a word of English.  It didn’t matter.  We were in the presents of Dr. Esaki.  Then we all went home excited to change the world.  Logic circuit designers knew the Esaki diode would lead to super high speed logic circuits.  RF engineers knew they would soon have breakthroughs in RF amplifiers.  About a month later the RF engineers were certain the Esaki diode would make a great logic circuit and the logic circuit engineers knew it would make a great RF amplifier.  It wasn’t much for RF and much less for logic circuits.  The tunnel diode has only two terminals so the input is also the output and vice versa.  The logic was performed by summing resistors.  Regular diodes required too much voltage for a tunnel diode.  The input resistors were just like the resistors the output drove for the next circuit.  I got a patent or maybe two on tunnel diode circuits but none were of any value.  The Wikipedia article on tunnel diodes claims they were used for switching circuits and high-speed counter circuits.  It provides no citation or source for these claims and the one reference to switching circuits, I don’t believe has any reference to tunnel diodes.  At any rate the tunnel diode is not likely related to the diode logic we are discussing.  It has had some success in RF amplifiers.  I designed a circuit using a tunnel diode to measure nanosecond voltage peaks to test magnetic cores for magnetic core memories.  Transistors were not fast enough.  Basically it is sometimes usable for a single stage or a part of another circuit.  I won’t say it was never used successfully as a logic device but I doubt if it was ever the only active device in the circuit.  It has twitched my interest again and I have an idea that probably won’t work but I might give it some thought.


 * With respect to diode speed, I will try to write a new section that deals with design with real diodes that addresses speed.Thingmaker (talk) 19:23, 26 August 2014 (UTC)
 * I first ran across a tunnel diode while sorting through a hobby pack of "50 parts swept off the floor for $1.79" - looking for good diodes, I tested one odd looking one and the TV chassis I had playing in a corner of the shop lost audio every time I put the meter leads on the diode. By random chance the lead lenghts were just right to make it oscillate. I'll have to check the old GE handbook, it might have some useful references for this article; I know it had a quartz clock where the divider stages were all tunnel diode multivibrators, but that's not the sort of diode logic that this article is about. Regular junction diodes have stored charge which has to be cleared to get them to switch - this is one limitation on speed. --Wtshymanski (talk) 01:28, 27 August 2014 (UTC)


 * I am aware of the stored charge but if it is working with a similarly made transistor inverter the transistor will not pass the glitch. Because of the amplification of the collector base capacitance the transistor is too slow even with fast circuits. If I am wrong where is the cure in real circuit designs? If you used old discrete silicon diodes with faster germanium transistors they may have had a problem but no one did that because the silicon diodes were expensive and slow.  In all the DTL circuits I designed and waa aware of I never experienced a problem with the recovery speed except with the very slow  selemium diodes with germanium junction transistors.Thingmaker (talk) 00:31, 28 August 2014 (UTC)

A diode switch
Does the 550 switching diodes for the clock in the Diode–transistor logic article prove that the PN diode is a switch? I have the explanation of how a diode switch works. Click on this User talk:Thingmaker. The simple PN diode chip puts integrated circuits to shame. Possibly instead of looking for published documentation of “someone” calling the diode a switch one might just read p-n junction or p-n diode and see if they can find how the pn diode is a switch.Thingmaker (talk) 20:40, 27 August 2014 (UTC)

DDL Logic?
Has anyone seen the recent work to create inverting diode logic by using an RF power supply component?

Notes on this: The logic circuits themselves are diodes with passives (resistors, capacitors and inductors). The circuits the person has created have an RF power supply that is built around a 74AC14 (Hex Schmitt inverter), but it's limited to the RF power supply. It should be possible to make such power supplies without the use of IC's, and possibly even without the use of transistors. May only be worth mentioning in the context that while inverting diode logic is apparently possible, it requires a lot more circuitry and an RF supply. Cefiar (talk) 03:14, 30 May 2016 (UTC)

I am tempted to change the first paragraph into two, with the following content:

In logic gates, logical functions are performed by parallel or series connected switches (such as relay contacts or insulated gate FETs like CMOS) controlled by logical inputs or parallel resistors or diodes which are passive components. Diode logic is implemented by diodes which exhibit low impedance when forward biased and a very high impedance when reverse biased.

There are mainly two kinds of diode logic gates - OR and AND. While it is possible to construct an inverting logic diode gate, it is fairly impractical and requires a large amount of extra circuitry apart from diodes and resistors (capacitors, inductors and an RF supply). The only other way to create a NOT or Invert function requires an active component such as a transistor. As such, this article will concentrate on basic diode logic.

Any objections? Ping Thingmaker & SpinningSpark. Feel free to make the change yourselves or rework the above to suit. Cefiar (talk) 05:32, 21 June 2016 (UTC)


 * I certainly do not think that this belongs in the lead. It is an obscure design and there is no evidence that this has any real world applications.  It appears to be purely part of a personal project.  I would also question the suitability of the source for Wikipedia use.  As far as I can see, the content is user-generated and thus comes under the heading of self-published.  From what we have at the moment, I think it is questionable whether this arrangement is notable enough to be on Wikipedia at all. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 14:31, 21 June 2016 (UTC)
 * This project is not diode logic, it is PIN diode logic. It would only belong here (and I think it does) within a section specifically on PIN diodes. Their use in logic is notable, particularly for airborne and phased array radar. In general diode logic it's just confusing though.
 * This isn't a project to make use of diode logic, it's specifically one to use PIN diodes as gain elements, and to use them in a context where they make no sense and where the RF is being supplied in order to feed the diodes, rather than PIN diodes being chosen because it's an RF problem. Andy Dingley (talk) 14:56, 21 June 2016 (UTC)
 * Sure, PIN diodes are widely used as RF switches, but are they ever used as elements in a logic circuit in a real application? <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 15:05, 21 June 2016 (UTC)
 * What's a "logic circuit", what's a CPU and what's just a couple of diodes? The most complex example I've seen myself has been address decoding and switching on phased array radar. Andy Dingley (talk) 17:54, 21 June 2016 (UTC)

Handmade transistors
Why did you remove the note about handmade transistors as unsuppoted? This book says that making handmade transistors ended in 1952, which means they were likely not used in the IBM 608 and the note was probably correct that they were used in the IBM 360 and 370 (and since that is the subject of the book, makes it even more likely). That has now left the article saying something almost certainly not true. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 19:03, 17 August 2019 (UTC)


 * I have trimmed this down further to hopefully eliminate the possibility of publishing something wrong. I removed the note about IBM 360 and 370 because there wasn't enough context to understand what that was trying to say and no reference cited to help me. I tried to clarify what was left and it is possible I changed meaning and made things worse. I have now removed all the details about components because of WP:V issues and because this is not important information in this context (would be good the articles about 608, 360 or 370). ~Kvng (talk) 16:18, 18 August 2019 (UTC)
 * Not so sure this doesn't belong. IBM handmaking transistors by modifying germanium diodes seems to still be within the context of "diode logic".  It should certainly be somewhere on Wikipedia and an overview article seems better to me than spreading the information over several individual products.  Besides, the source I found was not specific on what products it was used in.  We can infer that from the dates given, but that would be WP:SYNTH. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 12:12, 19 August 2019 (UTC)
 * I don't think we should restore what was there but I am not opposed to putting in something about this that's verifiable and that readers can understand. ~Kvng (talk) 14:16, 21 August 2019 (UTC)
 * Did you read my addition this morning? Are you ok with that? <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 16:03, 21 August 2019 (UTC)
 * That is clear and verifiable. It doesn't really make a connection to the section topic (Early diode logic with transistor inverter) but this section is already disjointed. Hopefully someone will be able to tie all this information into a coherent narrative at some point. ~Kvng (talk) 16:45, 24 August 2019 (UTC)