This DIY Project Using OpAmps is doing strange things. Fixing These Faults Is How I Learnt Repair
HTML-код
- Опубликовано: 18 сен 2024
- I found a shared Project on PCBWay to build a transformer dot convention tester. It didn't work at first and gave me a bit of a runaround to figure out why. In doing so I found some unexpected behaviour in OpAmp circuits. This video is a great way to learn how OpAmps work in real situations, and there are still some UNANSWERED QUESTIONS here for you guys to figure out. The last 20 minutes of the video features a 'live chat' session with Detlef and Myself where we discuss the circuit operation and possible ways to improve it.
This video is kindly sponsored by PCBway. Please give PCBWay a try for all your PCB needs. / pcbway
www.pcbway.com
If you use this link when creating a new PCBWay account and you will receive a free $5 discount coupon: www.pcbway.com...
Other creators and channels I work in collaboration with:
The Electronics Channel - live Q&A streaming sessions and more, every second Sunday.
/ @theelectronicschannel
Det Builds Stuff - / @detbuildsstuff8128 (Detlef)
Retro Upgrade - / @retroupgrade (Carlos)
Gran Canaria Uncovered - / @grancanariauncovered (with Detlef and Julie. Not electronics related)
Other PCBWay projects featured on LER
The 0-30V 0-10A PSU Project
Part 1 • How To Build A Variabl...
Part 2 • How To Build A Variabl...
Part 3 • Build this DIY 0-30V 0...
PCBWay Projects on Learn Electronics Repair
Amazing $1 Short Finder
• Convert Your MULTIMETE... Part 1
• The $1 DIY SHORT CIRCU... Part 2
• The Amazing $1 Short F... Upgraded Version
DIY Huntron Tracker / Octopus Curve Tracer
• How To Build the Amazi...
MilliOhm Meter
• Help!! My DIY Electron... Part 1
• Build A Cheap And Supr... Part 2
• Build A Low Cost Micro... Part 3
10KHz -255MHz RF Signal Generator
• Help!!! My Electronics... Part 1
• Build A Low Cost 10KHz... Part 2
300W Active Load
• Build A Low Cost DIY 3...
Amazing $1 ESR Meter
• Build This AMAZING $1 ...
Ring / Flyback Tester
• All You Need To Know A...
Amazing $1 Ring / Flyback Tester
• How To Find SHORTED TU...
Signal Tracer
• Build A Cheap DIY Sign...
Signal Injector In A Pen
• Build A Cheap DIY Sign... Part 1
• An Absolute Beginners ... Part 2
• An Absolute Beginners ... Part 3
Modified ATX PSU Analyzer
• This ATX PSU Tester An...
Single/Dual Op Amp Tester
• How To TEST OP AMPS Us...
Zener Diode & LED TEster
• How To Test Zener Diod...
TL494 Tester
• How to test TL494 / KA...
TL31 Tester
• How to test TL431 / KI...
Learn Electronics Repair is now on Discord! Come and join the fun, it's free.
/ discord
It takes a lot of time and effort to make RUclips videos. If you enjoy my videos or they helped you with your own repairs please consider subscribing to my channel or click LIKE
If you would like to help support this channel you could consider sending a Paypal donation or subscribing to my Patreon page. All monies received will go towards buying more items to repair and make interesting videos, or to improve my video recording equipment and will enable me to spend more time making better videos. If you subscribe to my Patreon you'll get my Whatsapp number and I'll give personal repair support
You can send donations via this link
www.paypal.com...
You can subscribe via this link
/ learnelectronicsrepair
Thank you for your support.
Richard
I really enjoyed this format ..one learns a huge amount trying things out and chattering and thinking 👍
Yes, unused inputs on CMOS device should be grounded, for the reason mentioned (oscillations) and to protect the inputs against static discharges. In case of a microcontroller you might ground the pins via a resistor as you could accidentally configure the pins as output. One should consult the datasheet of the microcontroller to know whether grounding of unused pins is advised.
yes i would love you and det do a micro controller version. i am new to this and would enjoy it. thanks for another great video!!
G,day from Sydney Australia. Congratulations to the designer and manufacturer of the PCB.
Learning outcomes
* alternative power attachment instead of banana plugs.
* reading a schematic
*Using a scope probe to detect DC and trace up movement of 5v
* regular frequency square (no noise)
Thank you for your time.
🌏🇭🇲
Before the modification pin 2 was GND and at pin 3 there was the voltage accross the primary coil (clamped to max 0.7V by the diodes) which is the output from NE555. Then you disconnected pin 2 from GND and connected it to the wiper of the potentiometer and the reason why then there was the signal at pin 2 is because of *virtual short* between opamp's inputs so the voltage at pin 2 must be the same as at the pin 1. Since opamp's input current is (almost) zero (input reseistance is almost infinite) the voltages at its inputs *when they can* be the same are always the same (check opamp virtual ground). By connecting the potentiometer (two resistors) you allowed for the voltages at both opamp's inputs to be the same and the signal at pin 2 was as expected, it was the same as the signal at pin 3. Just draw the schematics of the connection after you added the potentiometer and assume the voltages at pin 2 and pin 3 *must* be almost the same and that explains the behaviour.
You misunderstand the concept of a virtual short-circuit between the inputs. It exists *only* when an op amp is used closed-loop (i.e. with feedback) and then *only* in "small-signal" conditions. Feedback allows the output to drive the inverting input to a voltage very nearly equal to the voltage at the non-inverting input. With normal op amp circuits the feedback is local - directly from the output of the amplifier back to the inverting input. In many op amp applications the feedback path can be very "long" and involve a lot of other circuitry.
In the circuit at hand the amplifiers are being used open-loop, as comparators. There is no feedback so the virtual short-circuit does not exist. A large voltage differential between the inputs can exist. The specification for the LM324 allows the differential voltage to be as great as the total supply voltage.
@@d614gakadoug9 Everything I said is correct. Check the sentence where I said: "when they can". I didn't say there is a virtual ground in this case before the modification, I said to check virtual short and virtual ground topics to understand what happend *after* the modification. As soon as you disconnect the inverting input from GND and connect the trimmer all conditions for the opamp to keep both inputs at almost the same voltage are met and that explains the behaviour after the modification.
@@ChupoCro
No.
At 21:40 he begins to explain what he is doing with the potentiometer. It is connected between ground and +5 V with the wiper to pin 2 (inverting input of the op-amp). All that does is set a bias voltage on pin 2. There is still no feedback from output to the inverting input hence no virtual short circuit exists. The amp is still being run as a comparator and a large voltage differential between the input can exist.
@@d614gakadoug9 Well, as you can see, that isn't what happens. He tried to set the constant bias voltage on pin 2 but yet, the oscilloscope is showing a square wave on that pin. The opamp will do whatever it can to prevent voltage difference at its inputs and because of the two resistors formed by the trimmer it can do that.
@@ChupoCro
The ONLY way that the amp can created the "virtual short circuit" at the inputs is if there is feedback and there is zero evidence that there is in fact any feedback.
It would be helpful to have images of the PCB layout, but I assume that Richard is sufficiently competent to properly identify where Vcc and ground could be found on the PCB. [third edit - he confirmed with the scope that the pot was between Vcc & ground]. He did a very poor job of using the oscilloscope to gather useful information that might have helped to determine what was going on. The scope displays were fleeting and the scale factors were very poorly set.[edit] He also apparently had the oscilloscope probe ground connected to the negative side of the power supply instead of to the actual ground on the PCB.
[second edit] He also is known to buy parts from miscellaneous non-authorized vendors on the web, so who knows what the things that are supposed to be LM358s actually are. It seems that fraudulent and counterfeit parts abound.
If you have an input you don't need connect it to the correct logic level.
On an AND gate, unused inputs should be tied high. On an OR gate they should be tied low. Pins controlling sequential circuits such direct set or clear of a flip flop should be tied to their inactive level. If a whole gate is unused inputs can be tied either way as convenient - some could be high, some low..With CMOS you can simply make direct connections to the supply rails. Another alternative is to tie an unused input to another in the same gate - e.g. a three-input NAND gate might have inputs tied to signals A, A and B or A, B and B. In rare cases you might not want to do this because of fanout limitations.
With microcontroller pins it is best to tie them either way through a moderately high resistance so you don't get excessive current if your code is bad and sets a pin to output instead of input. The exception would be if you actually need the pulling resistor to source or sink for the benefit of the external circuit. Many micros have configurable passive pulling of pins. If you are doing a low power design be sure any external passive pulling isn't in conflict with internal passive pulling. It can be perfectly acceptable to configure an unused pin as an output if you can, as long as it is left open circuit externally or set to a level that doesn't conflict with any external pulling. If the pin function is set in code, do it early promptly after reset. Don't try to get by with "universal rules." Think about each and every pin and how it needs to behave with good code and what can go wrong with bad code. Beware of "oddball" pins such as those that can be used for an external crystal for the timebase oscillator or for I/O.
As other have alluded to, there are multiple issues with this design. Something to add (if it hasn't been already) is that if the input was a perfect square wave (which it isn't) and the output of the op amps was perfectly in phase or perfectly out of phase (highly unlikely), then the output of the XOR would be a clean high or low (no oscillation).
I'm with Det ground it, use good practise everytime coupling cap cost pence,
Heya, lol was watching this till the last 5 minutes or so yesterday wen your livestream started ( which was very fun again) with these kind of projects you learn so much from you guys talking and dicussing the working and schematic's of these projects thank you 2
If you are going to use a microcontroller you should use something very common and cheap. I suggest an Arduino Pro Mini which is available for about $2 and you can use the Arduino development environment (IDE) to program it using easy to understand code so that everyone can learn.
Hi g4z, currently we're set on an ATTINY85 - naked chip. We will program the chip using the Arduino IDE, so if you want to go Pro Mini, this should be easy to port.
@@DetBuildsStuff ok. btw if you are going to reply leave the auto-added name there because otherwise that person won't be notified. It was pure luck that I came back here to check another comment and saw this one.
@@g4z-kb7ct thanks - didn't know that!
Wouldn’t even an arduino pro mini be massive overkill?
@@jasonkary8431 thai's my take here, too. Interstingly, from a price tag perspective you don't save much. :)
this sort of content would be great for your sunday show
For CMOS logic, I have seen the unused gates on a quad 2-input IC actually affect the outputs used gate. These need to definitively pulled high or low (pick one, it does not matter) so that the entire package works properly. I believe for the 74HC series logic family, the datasheets are explicit on this. I did not happen to catch what family you are using.
Hey Richard , what a nice tool and project👍👍👍👍
I believe the ground on the - input of the opamp is a misprint on the print. Tie the leads either high or low fixes the output.
yes we want to see it.... Deth go and improve it 👍
Thanks Dutch - let's see how this goes. There already some ideas floating around - watch out for a "design" episode where we two stoodges hang around, curse a lot and drink beer to get this on a uPC :)
Just listened to the discussion about PCBs.
*NO! NO! NO! NO! NO! Do not fill up a board with copper pour/flood!.*
This is absolutely rank amateur stuff!
Large copper areas are very useful if applied properly but make a major mess if they are not. I've seen many instances of modules sold to the hobby market where copper pour for ground has been used and it has turned something that could be safe to something that is extremely dangerous. An example is the ACS series of hall effect current sensors. The sensors have been carefully designed so that they may be safely used for AC mains current measurement. Incompetent use of copper pour on modules utterly destroyed the safety spacing. Anyone using such modules risks death! Again, this is rank amateur stuff.
In other cases ground planes done with copper pour need to be carefully considered even when adequate safety spacing is used. Planes may need to be made into separate areas to be sure currents are localized. Incorrectly placed gaps in planes, as produced by normal traces, can force currents into the wrong place.
[ edit - addition ] Copper areas unconnected to ground can significantly increase capacitance between circuit nodes which can cause noise problems or even unstable behavior. In some circuitry even a grounded pour area must have voids added in specific areas to prevent creating unwanted capacitance to ground that can again circuit performance. This is especially true on multi-layer boards or just thin double-sided boards. Remember that in a properly designed circuit, ground and all of the power supplies are at the same AC potential once the frequency gets up reasonably high. That's never 100% true but it is an objective of good design. [end of edit]
Copper etched from PCBs is reclaimed by any competent PCB house. Even in China where environmental laws are extremely lax in general, copper is valuable enough to make it worth reclaiming.
Great video, I prefer raw vids. (Well obviously with the irrelevant cut out if you know what I mean)
Re-Pins, I agree, always have them set to a known state you can check.
I would be great if you could make this device into an Arduino Project. A nano if possible.
Hi Tim - Rich and my talked a bit more of the uPC version; I guess we'll go for a naked MCU; currently it looks like we're doing a ATTINY85. We discussed SMT versus Throughhole and our resoning was, that TH will be easyer if your transformer goes haywire and kills the MCU; harder to replace in SMT.
We'll probably do a "Design" Video as a follow up where you can follow these trains of thoughts, b'c this is a bit like having the voices in your head playing out loud :)
@@DetBuildsStuff Sounds good, the chip is the same family, when I prototype I tend to start on a nano module so there is plenty of elbow room for lazy coding, then when I am happy with the prototype, I then take a hatchet to the code (get rid of the Arduino architecture) so it will fit on something smaller.
I look forward to the project.
Cool idea.... Re the CMOS: why not dump the 555 and use a couple of the spare XORs to create the oscillator - though it'd probably need a buffer trans. Could the bizarre pin 2 thing be parasitic pickup from the transformer wires (although the pot doesn't seen to be high resistance enough to allow that...?
As for pin 2, it should be set to half of the supply voltage with a voltage divider. They try to use opamps as comparator but the inputs go to the limit of the supply voltage. Not a great design. The thing you see might be a result of putting a voltage on the inputs close to the supply voltage - which one should not do with this opamp. You should try to keep the inputs close to half of the supply voltage.
*As for pin 2, it should be set to half of the supply voltage with a voltage divider* We thought alike here, and then it worked. But can you explain what we saw on pin2 once we connected it to a simple voltage divider?
@@LearnElectronicsRepair I reckon if you added the decoupling caps mentioned and grounding unused inputs, it may well resolve this problem, unstable power rails. I've had many unpredictable results with cmos.
Indeed, and could be easily done for a quick check.@@mikepanchaud1
@@LearnElectronicsRepairyou see the defects because you are using the components (opamp) outside their spec (inputs too close to Vcc and ground). You could just leave the whole opamp out and couple the output of the 555 directly to the input of XOR gate. The opamp has no function.
@@kriswillems5661 *The opamp has no function* This is the conclusion we came to when we thought about the microcontroller version 😉
Temporary replace 555 with proper sq generator, they have so many odd output issues. add 1N cap to pin 2 to ground and remove XOR . Add decoupling in abundance. Reduce frequency to 400 hz or less. Measure it. I assume the unused XOR gates could be used to generate 50hz leaving gate open cct and the 3rd gate could square it up to 5 volts replacing the 555? Might be prone to trace and internal cross talk tho?
I'm tending to believe that people don't make the gerbers or schematic available because they think it will stop people copying it. What they don't understand is anyone with the PCB can reverse it easily and make a copy and/or improve it and re-release it (I've been doing that for many years). pcbway should make it mandatory for shared project uploads to include at least the schematic and bill-of-materials because not doing that just makes everyone involved look really stupid, including pcbway.
I agree with all of this. Responsible authors sharing projects should add all the files, schematic, gerber and BOM.
What Richard said :)
1. The op-amps are being used as comparators.... an ideal op-amp works as a comparator but not in reality. There should be some feedback resistors to limit gain and allow the op-amp to function and, preferably, input resistors. Using a comparator IC would be better.
2. Remember that the input to an op-amp is CURRENT not voltage and the op-amp amplifies current. If you think carefully you will see why the resistors are needed.
3. When ideal op-amps are analysed the load is generally resistive. Higher level electronic analysis is needed for the heavily INDUCTIVE load depicted here. The output to input current phase change must be considered and extra components added for correct operation.
4. The transformer secondary is clamped to a short-circuit by the back-to-back diodes. Any transformer that has a low turns ratio will suffer a collapsed magnetic field that will put huge current spikes on the op-amp output..... I would add a resistor between the op-amp output and the transformer at the very least but preferably a separate driver.
5. Decouple the power supplies at the IC pins - remember that putting a scope on a circuit changes that circuit especially those lacking decoupling.
6. As other have said, always ground unused inputs.
2. Remember that the input to an op-amp is CURRENT not voltage and the op-amp amplifies current. If you think carefully you will see why the resistors are needed.
Please clarify. Op Amps are supposed to by Hi Z inputs, in the ideal op amp model the input resistance is infinite, therefore it can't be amplifying current as there is no input current.
@@LearnElectronicsRepair Phew, that's not easy in a fee lines of comment..... but
You are using LM358.... Google the Texas Instruments datasheet and the internal schematic is shown there.
You can see that each input to the chip is connected to a standard PNP darlington pair of bipolar junction transistors.
RUclips has videos on darlington pairs and their beta current gain. This explains the input to the device.
Op-amps usually have resistors around them which allows us to convert currents to voltage and since the currents into the op-amp are small we can neglect them and use pseudo voltage analysis.
However, hee we have no feedback resistors so we must analyse with current.
A 393 Voltage Comparator IC is a Voltage device (where feedback is handled internally) and is pin for pin compatible with the 358. I'd suggest swapping one into the circuit to make analysis easier. You will need a pull-up resistor as the output of the 393 is open collector.
You may find a 393 works better without the input clamp diodes.
@@LearnElectronicsRepair Regarding the ideal op-amp with completely zero input current.... the ideal op-amp also has completely zero output impedance.
If the real op-amp did actually have such characteristics we would be able to remove the op-amp and it would still work ! Obviously this is wrong so the real op-amp must have a very small input current and a very small output impedance.
There are other parameters that exist in real life but are ignored for simple analysis; for instance, an ideal op-amp has zero phase shift between input and output allowing it to work at all frequencies, a real op-amp does have a phase shift which means it can only handle lower frequencies. This phase shift may also cause oscillations etc.
If you find some op-amp based audio pre-amplifier circuits you could play about with this 👍
@@LearnElectronicsRepair Sorry if I'm droning on but one more thought occurs to me that may help the analysis.
When the 555 output is low at start up the voltage on pin 3 of A1 will also be low. No current will flow in the primary.
When the 555 output goes high the voltage on pin 3 of A1 will stay low initially due to the inductive primary of the transformer. As time goes by the field in the transformer will increase and the voltage on pin 3 of A1 will ramp up too. It will be clamped at 0.6volts. There will be a current flowing in the transformer primary.
When the 555 goes low the current into the transformer primary immediately stops. The inductor tries to maintain current and the voltage across the transformer primary will reverse, ie it was 0.6volts when 555 was high so it will go to minus 0.6volts. Thus the voltage on pin 3 of A1 will be -0.6volts.
This will be enough to switch the comparator output state.
As time goes by the magnetic field in the transformer primary will decay and the voltage will fall back to zero volts.
An op-amp without any feedback resistors and with its inputs both close to zero volts will be unstable so the output will be unknown or possibly oscillating.
Therefore it's better to have feedback around the op-amp to provide hysteresis ir use a comparator IC with built in hysteresis.
edited to fix type number for op amp (I had used number for quad - LM324)
1, Op amps can be used as comparators but there are many things to consider. With the LM358 the biggest issue is that it is quite slow. At low frequency that doesn't matter. [edit] I have used the LM324 (quad version of same amp) as a comparator in an industrial product where it has functioned entirely satisfactorily for decades. I wouldn't even consider it if I needed something fast.
2. Conventional op amps work on the basis of voltage differential between the inputs. There _is_ some current flow but it is small. For the LM358 at room temperature it is typically -10 nanoamps. It is specified as negative because conventional current (i.e.flowing from positive to negative) flows out of the inputs. For many applications with low source impedance the input bias current can be ignored. In precision circuitry it must be considered carefully because it introduces an error. Some bipolar op amps have internal bias current "cancellation" circuitry. JFET op amps have inherently much lower input bias current and CMOS amps lower still.
3. The op amps in the circuit at hand play no role in driving the transformer.
4. No.
5. The circuit should have better decoupling, but poor decoupling will show up around the time of the transitions, not "steady state" so it will have no functional effect on this very slow circuit. Loading by the scope probe will be unimportant in this circuit because nothing is high impedance.
6.No, don't ground unused inputs. Connect them to the appropriate logic level. For completely unused gates inputs can be tied either high or low. Different input of the same unused gate can be tied differently.
[edit - addition] Do not simply ground inputs of unused op amps. You want to keep unused amps in their linear region to minimize wasted power. There are several satisfactory ways to deal with unused inputs but this is well documented elsewhere and there are too many ways to try to describe here.
Deff have a go
Hi Malcom - yeah, it's in the making. We talked today about SMT vs Troughhole and ended up with the latter; we think we can do this with a pretty low component count and no special (=expensive) parts. There probably will be a "design" episode where you can follow us cobbeling this thing together.
The appearance of the signal at pin 2 of the op amp is strange. It could have been helpful to actually have good information from use of the oscilloscope instead of the crude yes/no-signal observations. Knowing exact voltages at both inputs when the signal was showing up on pin 2 could have provided more clues. The best way is to use two channels simultaneously, with the zero volt DC baseline set to the same place on the screen - which would NOT be the centre of the screen.
I think I heard some mention of small amplitude signal at the output of that amp even when it wasn't working. I think I also heard that the ground of the scope was connected to the power supply ground. If that is the ground on the PCB that's OK. If it was at the other end of wires connecting the PCB to a bench supply it most certainly is not OK.
The input bias current of the LM324 is on the order of -10 nA typical at "room temperature" (negative indicating conventional current flows out of the input; there must be a DC path to the negative supply rail of the amp through which that bias current can flow). Because the current is very small the pot should have totally "swamped" it unless that was some very bizarre ultrahigh resistance pot.
Presumably the person who "designed" the circuit originally made it work reliably. It might, even with pin 2 grounded, if the input offset voltage of the op amp happened to be favorable. Even that seems a bit unlikely because the output of a bipolar 555 doesn't swing very close to ground. A CMOS version would swing right to zero volts if it were sinking no or very little current.
The only reason I can see for using the amplifier/comparator that is driven directly by the 555 is to equalize the delays, at least to some extent. The LM324 when used as a comparator is slow! That's perfectly OK for lots of things but could cause some "spikes" at the XOR output with this circuit. Those huge caps are presumably there to fix some sort of delay/spike problems that do exist.
Yes
Looking at it and thinking... it CAN'T work.
Pin 3 is being driven by the 555 and unless the 555 can pull right down to 0V, will always have a little +V, while pin 2 is at 0V. I don't know if the LM358 is even capable of working with inputs around the single ended 0V. I'd forget that half of the op amp, or use a single, and directly drive that XOR input from the 555.
And also ensure that the unused CMOS inputs are tied, or parallel a couple of the XOR gates to ensure enough LED drive
Right, the - input must be connected via another 100 ohm resistor to ground, not directly to ground. The + input must be set to half the supply voltage via a divider. The opamp is used outside its spec.
The output of a CMOS 555 will swing to zero volts if it isn't sinking more than a very small amount of current. The bipolar version won't.
The input common mode voltage range for the LM324 does include the negative rail and it would be OK with a few tens of millivolts below the negative rail. Anything more negative than just a very small amount below the negative rail and you can get phase reversal. Even with that and a CMOS 555 you are up against the input offset voltage of the amp which can be either positive or negative. The inverting input of the upper amp needs to be biased at least a quarter of a volt or so above the worst-case maximum LOW voltage from the 555. You could use the same network as used for the other amp. Half of Vcc isn't in any way optimal, but it's OK. Bias voltages like that should usually be filtered.
@@kriswillems5661
I don't see what those things would accomplish.
Putting a 100 ohm resistor between the inverting input and ground will raise the voltage of the pin by about a microvolt due to the input bias current of the amp. That is still vastly below the expected input offset voltage.
Biasing the non-inverting to half the supply voltage when it is being driven by a low-impedance source does nothing of value.
@@d614gakadoug9 you are right. Is there any reason the 555 timer can't be directly coupled to the XOR gate? I don't know why they use an opamp to start with.
@@d614gakadoug9 or how about setting both inputs via 10k resistor dividers to half of the supply voltage. And then tie the 555 via a 10k to one of the inputs. That way you stay away from Vcc and ground.
Can this device be used to check for shorted transformers
No point, you have a multimeter.
Please create your alternate idea … I really enjoyed the dialogue between you both …
go for it
Cut the blue wire to pin two and put the scope on blue wire and then on pin 2 to see where the signal is coming from.
The 555 has a DC pedestal !
Normally 1 V or so ( Low is not 0 V ) .
A quick fix is jack it up the op amp !
2 diodes in power ground ( pin 4 ) .
If that is all you do the inputs of the amp will be driven or pulled below the amp's negative supply rail by an excessive amount which leads to phase reversal at best and damage at worst.
29:26 Do you not have a toggle switch handy?
The microcontroller version will be interesting but it's kind of like shooting a mouse with a cannon.
K.I.S.S. engineering (Keep It Simple Stupid) is beautiful in it's simplicity when it works.
Since a transformer is an AC device I would have thought a simple AC output wall wart power supply would have been a starting point. Those 555 chips are cheaper though.
The board design also could use some holes for mounting it inside an enclosure.
" ... I would have thought a simple AC output wall wart power supply would have been a starting point"
Huh? That makes no sense? Starting point for what?
@@d614gakadoug9 A starting point to acquire a stable waveform to energize a transformer. The whole point of the 555 section of the circuit is to mimic an AC waveform to then use to excite the windings of a transformer.
@@tenlittleindians
That would be OK for iron core transformers intended for AC mains use but completely useless for high frequency transformers which typically have such low winding inductance that they'd look like a short circuit at 50 or 60 Hz.
I'm not entirely convinced that any single frequency is useful for "all" common transformer types.
You had to order at least 5 of the PCBs, what do you do with the rest? Isn't that a lot of waste for 4 of them and just for the landfill? Would you ship me one if I pay you the postage? :I live in Germany. :-)
Is this why he changed the voltage to 3.3volts ,to decrease the signal
This thing is just a toy ….
But toys can be useful !
Diodes across the inputs are for voltage limiting ( IN400x ) so a 1 : 100 transformer shouldn't kill the IC !
Adding a inline resister might be a good idea .
As for the 74 logic you just don’t need it .
Just have 1 led + load resistor + diode & a switch flipping one coil .
We (Det and myself) are going to try a microcontroller based version that has less components and is cheaper than this one. Toys can be fun, and also make useful teaching aids sometimes. Certainly this one created a lot of interesting topics 🙂
Signal diodes like 1N4148 are much better than 1N400x types for protection of the inputs. Their "forward recovery time" (rarely specified, but tends to go with reverse recovery time though much shorter) is very substantially less. They are entirely adequate for the currents that might be involved. The LM358 actually won't be damaged with differential input voltage up to the maximum power supply voltage allowable, even when operating on a much lower supply voltage. That is far from universally true, so care is required with other comparators or amps. Series resistors are always a good idea when compatible with other design objectives. You have to pay attention to the input bias current and input offset current of the amp or comparator and how that plays with series resistors.
If you are going to build a dedicated test circuit, why not make it do as much as it can without the need for fiddling with switches?
@@d614gakadoug9
A switch or one more IC ?
I go with the switch or just move the primary connections until I get the led …..LOL
@@gd2329j
A switch of decent quality is very much more expensive than the IC and very much less reliable. If you're going to go to the bother of designing a PCB you might as well do the circuit right and make use convenient.
The only utility I can see to making a circuit like this is for production testing where time is important and the less futzing around the better. I've designed and prototyped many transformers for SMPS and simply used existing equipment to verify phasing where necessary - which was extremely rare even when phasing was important (forward and flyback converters and gate drive transformers).
This is your typical amateur project..... using kicad with very little to no knowledge of pcb layout and part selection, a quick autoroute, quick upload to pcbway shared projects thinking they will make millions and they upload nothing else because they are amateurs and have never designed or made anything professionally and never had to build something from scratch after years of doing other stuff (i.e. years later and forgotten everything). Most of the stuff on pcbway shared projects is like that which makes most of that stuff pretty useless. Over the last 10+ years I've been taking similar amateur projects that interest me and improving them with cost-reduced part selection (one example someone made a project using a ten dollar sdcard slot LOL!) and/or reversing them and releasing schematics and gerbers publicly so that these projects can live on and not just die when the original person gets bored and stops making it available which is very sad and very common unfortunately, especially with retro-related projects.
You should be commended for your input to the hobbyist community. Keep up the good work. I take a slightly different approach as this is primarily a repair based channel, but building these amateur projects and then getting them to work or finding out why they can't work is also a great way gain electronics repair skills. With Det's involvement we are also now building improved versions specifically trying to make reduced component count versions using a bare microcontroller. it seems the viewers on both channels like this content. Det's channel @detbuildsstuff8128
oke ummmm you got the pot on vcc and gnd the middle wire to pin2 so I think you pick up some how the signal comming on pin3 are you sure you are using the exact opamp as it is calling for ?
Yes LM358
I have to agree with Deth should be grounded
Det should be grounded?? Why? What did I do this time??
This seems like a long way to go to create a square wave signal generator.
Test without cemos in
I could do this with an arduino