Alimentator PWM analogic

[mpursu]

Schema propusa de mine pentru un alimentator pentru trenuletele analogice este una foarte simpla, dar de efect:

Intentionat nu am desenat aici schema de alimentare separata a circuitului integrat CD4093 cu tensiune stabilizata de 9V, deoarece aici se pot utiliza fie montaje discrete cu diode Zener, fie stabilizatoare integrate gen LM7809. Borna VDD se conecteaza la o sursa de tensiune filtrata si nestabilizata de 17V, de unde isi ia tensiune stabilizatorul pentru CD4093.

 

[mpursu]

VDD este tensiunea pozitiva de alimentare a intregului montaj, de 16 - 17 V. In functie de pozitia cursorului potentiometrului, montajul emite impulsuri de 12 - 14 V (din VDD se scade tensiunea de saturatie a tranzistorului, circa 2-3 V) cu factorul de umplere intre 2% - 98%, la frecventa de circa 25 kHz, insesizabila pentru auz. In pozitia "min", montajul emite impulsuri foarte inguste, care aprind farurile LED, dar inca nu misca motoarele. Miscand cursorul spre pozitia "max", latimea impulsurilor creste, iar motoarele incep sa se roteasca cu viteze din ce in ce mai mari. Sensul de deplasare poate fi inversat cu ajutorul comutatorului SW1 - preferabil NU in timpul mersului :-?. Acest montaj ofera posibilitatea mersului stabil al locomotivelor si la viteze foarte mici, spre deosebire de cazul alimentarii trenului cu o simpla tensiune continua de valoare variabila, cand reducerea tensiunii provoaca reducerea vitezei, dar si a cuplului motor, mersul locomotivelor la viteze foarte mici devenind problematic. Datorita faptului ca farurile LED nu au inertie termica, lumina acestora este perfect vizibila si atunci cand trenurile stau pe loc, intensitatea luminoasa crescand usor odata cu cresterea vitezei de deplasare.
Ideea nu este noua (practic am dezvoltat un montaj din cartea "301 montaje electronice"), iar alimentarea PWM a motoarelor de curent continuu este des utilizata in automatizari, modele telecomandate etc., existand chiar si circuite integrate specializate pentru asa ceva.
Acest alimentator PWM nu este compatibil cu locomotivele digitale, chiar daca acestea recunosc si alimentarea in curent continuu, fiindca decodoarele sunt "dezorientate" de impulsurile primite, pe care nu mai stiu cum sa le interpreteze. Aceasta situatie se poate evita prin coborarea frecventei impulsurilor PWM pana la 100Hz.

 

[mpursu]

If needed, I can translate all these in English

 

[mpursu]

Montajul meu este protejat la scurtcircuit - mai degraba la supracurent - prin grupul Q2-R3. Anume, daca valoarea curentului prin rezistenta R3 ajunge la 1A, atunci caderea de tensiune de la bornele ei ajunge la 0.68V, ceea ce deschide tranzistorul Q2, care pune la masa baza lui Q1, blocandu-l. Astfel, curentul prin tranzistorul Q1 si circuitul mini-feroviar este limitat la 1A. Pentru curenti admisibili de valori mai ridicate, se dimensioneaza corespunzator rezistenta R3, care trebuie sa si aiba o putere adecvata.

 

[PrinzEugen]

i use google translate from romanian to english...

summary
if i understood this controler is


- overcurrent above 1a and short circuit protected,


- it can be used wit reverse polarity kipp switch for change direction


- also no heatsink is needed for BD651 and LM7809, and all capacitors are ceramic, all resistors with at least 0.4w wattage


-regulation from 0 ? or 7v as starting regulation?

 

[mpursu]

Close enough 8)
This controller feeds the analog railway system with PWM impulses, with fixed amplitude (12 - 14V) and variable width. This means that the output voltage goes from 0 to 12-14V and back to 0 with a 25 kHz frequency. The mean output voltage can only be approximated with an analog voltmeter, which will go from close_to_zero to close_to_maximum_voltage when the potentiometer is rotated from "min" to "max".
Please check again very thoroughly all the components and their connections, then try again to operate the device. If it still doesn't work, tell me and I will guide you step by step, as I did with our colleague BDA.

 

[mpursu]

Acesta este un montaj foarte simplu care ofera impulsuri PWM cu frecventa fixa si latime variabila, daca nu chiar cel mai simplu. Cu toate acestea, conexiunile imperfecte, cablurile intrerupte sau componentele neverificate pot sa cauzeze nefunctionarea sa, deci... stiti ce aveti de facut

This is a very simple device which gives PWM pulses with fixed frequency and adjustable width, if not the simplest. Even though, wrong connections, interrupted cables or not checked components may prevent it from correct operation, so... you know what to do

 

[mpursu]

Iata aici partea de comanda a alimentatorului meu dublu

In dreapta se afla primul alimentator, realizat cu portile 1 si 4, iar in stanga se afla al doilea alimentator, realizat cu portile 2 si 3. Portile 1 si 2 sunt oscilatoarele PWM propriu-zise, iar iesirile lor sunt conectate la intrarile portilor 4, respectiv 3. Firul verde si firul albastru, care pleaca de la pinii 11, respectiv 10, duc semnalele PWM spre tranzistoarele finale. Cele doua grupuri de cate 3rei fire duc spre cele doua potentiometre de 100kohmi.
Intre timp am trecut pe digital, deci nu m-am mai ocupat de proiectarea & realizarea unui cablaj imprimat pentru acest dispozitiv. Totusi, nu l-am demontat si l-am pastrat asa cum era.

 

[mpursu]

There is the command part of my double PWM analog controller

The first controller is on the right side, made of gates 1 and 4, and the second controller is on the left side, made of gates 2 and 3. The gates 1 and 2 are the PWM oscillators, and their outputs are connected to the inputs of gates 4, respectively 3. The green and blue wires connected to pins 11 respectively 10 drive the PWM signals to the final power transistors. The two groups of three wires each go to the 100k potentiometers.
In the mean time I passed to the digital system, so I did not design any PCB for this device. However, I did not take it apart and I kept it as it was.

 

[PrinzEugen]

Here it is revisited scheme for PWM analog controller on split strip-board

I have checked board for 7 time literary. All connections on board are like on scheme.

Potentiometer doesn't working, stuck to 14v as input. No min or max.

Readings are 14v on output. Led dont lit. Motor dead on 14 DC voltage
No cold solder joints also.
Output and input are not short connected.
Positive input is on + rail (checked). Negative input is on - rail




5-6-7-8-9 cd4093 to ground . 4 and 10 free pins


Q1: BD651 transistor
Q2: BC547
C1: 1nF capacitor
C2: 0.33 microF capacitor
C3: 0.1 microF capacitor
D1 and D2: 1N4148 diodes
D3: 1N4007 general usage diode
R1: 2.2 KOhm resistor
R2: 2.7 KOhm resistor
R3: 0.68 Ohm resistor
R4: 1 KOhm resistor
P1: 100 KOhm potentiometer linear mono

Maybe pot is wrong connected ?...i dunno...capacitors dont have polarity also.

 

[mpursu]

Here we go again
So you placed an analog locomotive on the railway and you connected OUT+ and OUT- to the rails. If the analogic voltmeter connected to the rails shows 14V and the locomotive motor and lights don't work, then there might be some problems with the rails or with the locomotive itself. At 14V any locomotive should run considerably fast with the lights brightly lit. Please repeat the test with another locomotive or with a simple electric 12V motor connected directly to OUT+ and OUT-. If the new locomotive / motor will run / rotate, then the first locomotive has contact problems which have to be solved.
If the voltmeter does not move from 14V when the potentiometer is rotated from "min" to "max"... that's another story, which we will solve after we make sure that the rails & locomotive are OK.

 

[PrinzEugen]

tested with 2 mehano locmotives and 3 piko lokomotives...nothing. POT is stuck to 14v. rails are ok...all checked...even light bulb on 12v wont lit up


also tested with bare wires with no rails...wont move a bit.... :lautar:


BDA manage to build controller


lucky guy hehehe

 

[mpursu]

So the voltmeter firmly connected to the rails shows 14V and all locomotives placed on the rails do nothing, at any position of the potentiometer?!? I have never met such a situation
What kind of voltmeter do you use, digital or analogic? If the voltmeter is digital, shows 14V and the locomotives do nothing, it might mean that the PWM oscillator gives VERY thin impulses, which cannot rotate their motors or light their filament bulbs. It would be interesting to place on the rails a locomotive with LED lights :idea:
BDA did also have some problems, but they were caused by faulty wires. Thay's why I was insisting to check ALL connections
I have build several such PWM controllers, which worked perfectly from the very first time, with no additional adjustments or repairs, so I am very surprised to find out that other colleagues have trouble building them too.

 

[mpursu]

For R3 (0.68 ohm) you should use a power resistor with at least 1-2W, not a subminiature resistor, because it is traversed by all the currents of the trains on the railway and might get burnt by overheating.

 

[PrinzEugen]

i will try with larger resistor...multimeter is digital..fluke

i suppose is small current or impusles...but why pot is stuck to 14v?

 

[mpursu]

Please explain "pot is stuck to 14V" :fluierat:
What do you mean? The voltmeter across the rails shows always 14V with the potentiometer at any position?
Let's try something else, in order to figure out what is happening. The digital voltmeters try to show the instantaneous value of the measured voltage, unlike the analog ones, which show an average value of the measured voltage. In the case of this PWM controller, the indication of an analog voltmeter will give us some indication about the speed of a motor driven by the PWM signal, and the indication of a digital voltmeter will somehow oscillate around the maximum voltage of the PWM signal. So it may seem logical that the digital voltmeter always shows something like 14V, because this is the amplitude of the PWM signal.

 

[PrinzEugen]

is says 14.3v at min and 15.5v at max...even led on 2v with 20mA of consumption wouldnt light on on output without resistor???, nothing .??? like no current or current is so small

interesting thing is...i must say that when i on my first scheme wrong connected base of bc547 and bd657 a get 7v at min and 14v at max pot. and of curse no current at all.

NOTHING IS HEATING UP OF ELEMENTS> EVEN IVE JUST CHANGED NEW ONE BC547 AND STILL SAME. :?:

i must go to CERN to solve this... :lol:

 

[mpursu]

CERN indeed
An oscilloscope would be very useful here, to know for sure what's there. If you have one (or borrow), try to visualize the PWM signals at pin 3+12+13 (all connected together) and at pin 11, at different positions of the potentiometer cursor. At pin 3+12+13 you should see the pulses widest at "min" and thinnest at "max", and at pin 11 you should see the pulses thinnest at "min" and widest at "max". If you cannot see any pulses, but only continuous logic 1 or logic 0 levels, then the CD4093 does not oscillate and you should check once again the connections or even change the CD4093 itself.

 

[PrinzEugen]

cant find osciloscope at all..i have workers all around me..drunk ones...ive rather go buy new IC or ticket to Geneva and visit black hole....

 

[mpursu]

First of all we have to solve the mistery of the 14V output that does nothing to any locomotive, motor or even LED without any series resistor.
I simply cannot understand how the output voltmeter shows 14V, but nothing works if connected there. What happens to the output voltage when you connect something, with the voltmeter already there? I really must see that.