# 555 ramp generator

Discussion in 'General Electronics Chat' started by Thevenin's Planet, Aug 4, 2012.

1. ### Thevenin's Planet Thread Starter Active Member

Nov 14, 2008
183
1
http://www.allaboutcircuits.com/vol_6/chpt_6/8.html

The above is concerning these five questions.
#1How do you calculate the frequency for the above circuit at the above E-book site?
#2, what is the difference between sawtooth wave and ramp wave?
#3 The Television begins at the left and move to right but the oscilloscope don't assume,Why ?
#4 In regard to the television pixels,do they enter thru the cathode of CRT?
#5 Is the Scanning,that is,Horizontal and vertical plates on the CRT tube automatic or fixed procedure that always happen when a signal is not there?, gray scale,not color.

Last edited by a moderator: Aug 6, 2012
2. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
1. The frequency is going to be very close to 1/(.7RC), what it exact formula is I'm not sure.

2. Saw tooth waveforms have exponential functions, while ramp is a variation of a triangle, linear.

3. Sweep on both TVs and Oscopes both go from left to right.

4. Old style NTSC signals do not have pixels, the new system is digital, and does not use a classic sweep. Not sure of the details of it though. You are mixing CRTs vs LCD, they do not mix well. Pick one.

5. Scanning is constant. Intensity is variable.

Thevenin's Planet likes this.

Nov 14, 2008
183
1

4. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
A CRT uses a scan, and NTSC. A LCD is not a CRT, and does not follow the same rules. It is random access, where as a CRT is vector driven, as in an old oscope. You are mixing apples and oranges.

Exponential, as in RC charge curve.

555 Hysteretic Oscillator

Now compare that waveform with what is used for this function...

Pulse Width Modulation

5. ### vk6zgo Active Member

Jul 21, 2012
677
85
(1) What Bill_Marsden said.

(2)The E-book has it wrong--out in the real world,where real Techs live,a Sawtooth is a Ramp.
A Ramp in not necessarily strictly speaking,a Sawtooth,as it may not "flyback" instantly,but may remain at zero volts for some time prior to beginning its amplitude increase.
Both terms are used fairly synonymously in the field, though.
The other waveform shown is an exponential waveform.

(3)In normal use,an analog Oscilloscope beam moves from left to right ,just like a TV( actually it also has to move right to left during the retrace time) ,but there is no raster display--the vertical deflection depends on the signal you are looking at.
In XY mode,there is no retrace as such,& the beam moves both ways in an oscilloscope.

(4)This is a silly question.
Despite the misuse of the term "pixel" by many people,it is based on the words "picture element",& yes,CRT displays do have pixels.

The smallest picture element a raster display TV can present is dependent
on the vertical & horizontal resolution of the system,which,in turn depends on the number of active scanning lines(vertical res),& on the number of resolution lines (horizontal res).
The number of resolution lines which an observer can see depends upon the frequency response of the TV system,around 5MHz for PAL 625.
These two factors,limit the size of the smallest picture element (Pixel) which may be observed.
The idea of resolution lines predates TV,& was used in movie & still photography.
Of course,in those media,there are no scan lines,& the horizontal & vertical resolution are identical.

As the video signal is fed into the tube cathode,you could say that component of a pixel does come in via the cathode,but as I said,it is silly!

(5) Yes,of course the horizontal & vertical scans operate whether or not a video signal is present.
This happens in any CRT TV ,colour or B&W,
Some TVs cut off the tube beam when there is no video present.but that does not affect the scans.
When a composite video signal is present,the syncs are stripped off & fed to the horizontal & vertical circuits,where they are used as frequency & phase references for the respective scans.

PS: The waveform Bill uses in the PWM reference is a "triangle wave",which is like a sawtooth,except that both rise & fall times are the same.I suppose you could call it a special case of a "ramp" or a "sawtooth".

Last edited: Aug 7, 2012
Thevenin's Planet likes this.
6. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
2> Been doing this for 40 years, a ramp is exponential, and a triangle/ramp is linear. I have never seen it any other way. I agree with the book, which was written by a different person, another professional and a professor. Who is more likely to be wrong here?

The reason the distinction exists is because a sawtooth is frequently used as a triangle, with diminished results (class D amps will suffer some minor distortion, for example). With one word it is made perfectly clear which type of waveform you are dealing with, a mark of many professions.

Thevenin's Planet likes this.
7. ### vk6zgo Active Member

Jul 21, 2012
677
85
Again,out in the real world,the terms ramp & sawtooth are normally used as synonyms,& certainly in analog circuitry,designers go to great pains to linearize signals from multivibrators & the like,which can be more exponential in their raw form.

Triangle waveforms are very seldom used in linear circuitry,whereas,the terminology "sawtooth" has been around for the best part of a century.

I am neither an EE nor a Professor,but I have also been in this game for over 40 years.
It seems I have to forget all those references which I have read (in books by "Professors" & others),to sawtooth & ramp waveforms being linear, because someone disagrees on a website.

8. ### Bernard AAC Fanatic!

Aug 7, 2008
4,240
414
I find it best, if it makes any difference, to referr to ramps as linear or non-linear. I think i've bee at it around 67 years.

9. ### Thevenin's Planet Thread Starter Active Member

Nov 14, 2008
183
1

Last edited: Aug 8, 2012
10. ### vk6zgo Active Member

Jul 21, 2012
677
85
Feeling a bit hypocritical after my crack about "someone on a website",I "Googled" for "sawtooth waveform" & found numerous discussions & drawings of such a waveform.
All of them supported my contention that a "sawtooth" waveform is one where a linear change in time corresponds to a linear change in voltage.

I then "Googled" for "ramp waveform",& all the same sites came up,plus a few others.
All identify a "ramp" as being a waveform where a linear change in time corresponds to a linear change in voltage.(same description as a "sawtooth").

The normal charging waveform of the C in the CR network controlling a 555 is not linear,& therefore not a sawtooth.

The "constant current" signal is linear,& hence,a sawtooth & a ramp.

Thevenin's Planet likes this.
11. ### Thevenin's Planet Thread Starter Active Member

Nov 14, 2008
183
1
I reread the E-Book on 555 ramp generator,and discovered that it explained the use of the two PNP transistors in regard to getting the exponential wave form linear.I gonna use that config to achieve a horizontal and vertical oscillator to control the T.V. scanning plates for a oscillator.The approximation
equation I will try. Thanks again for the input.

12. ### #12 Expert

Nov 30, 2010
16,703
7,339
Just a cute little drawing that says a constant current into a capacitor creates a linear voltage ramp.

File size:
5.4 KB
Views:
64
13. ### Thevenin's Planet Thread Starter Active Member

Nov 14, 2008
183
1
(A)I try to get the circuit working but could not for some reason which is not misconection of any components.(B)Thus I constructed it without the constant current config.(C) The result were,measured the voltage across the capactor with analog meter that gave a alternating voltage of 2 volts to 1.8 volts.(D) Vd1(anode to Vsupply and cathod to output #3 pin) has an "On" period that seems to be two time as short as the "Off" period.(E)Can't sense the On and Off voltage with the analog meter but it rise to +1.4 volt and stabize there.(F)It was observed thru the D. C. current mode about 1 microamps,alternating. (G)The digital meter was random with the highest about 12 microamps.(H) D2(anode connected to pin #3 and anode to ground) for some strange reason continously glowed On.(I)I used a 33 microfard capacitor. (J)I don't know if the 470 mF worked,it didn't with constant current.(K)What happenning with the circuit without the constant current config?

14. ### Papabravo Expert

Feb 24, 2006
10,340
1,850
If we are talking voltage waveforms, I have always understood that a ramp is the integral of a step. It is a "linear" and not an "exponential" waveform. I've been doing this stuff for half a century and I have three Engineering degrees. I can also play the "who is more likely to be right" card. IMHO that is actually a fairly weak argument.

The only other explanation I can imagine is a hidden or missing context that allows both views to be correct.

15. ### Thevenin's Planet Thread Starter Active Member

Nov 14, 2008
183
1
Well in regard to the my previous thread it seems that somewhere in the process an error happen.I redone the experiment that was in the E-Book finally getting it to work with the 470 micro farad connected to the constant current config.The odd thing is that the current thru the controlled transistor (the one connected to Tp #1 and Tp #2) is that the meter measured 57 micro amperes to 58 micro-amperes.The current climbs very quickly to 58 uf,then reverse slowly back to 57 uf and repeat the process.D1 (anode to the positive rail) flashes quickly as capacitor is discharging.I expected that the current thru the constant current config would be much higher then micro amperes, an increase in Beta.Do the constant current config is less than 1. No amplification?

16. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
Not sure what you are asking?

The current is set by the resistor in the current mirror.

17. ### Thevenin's Planet Thread Starter Active Member

Nov 14, 2008
183
1

I assume that you are referring to the 100K ? What I am asking, do this circuit,that is the mirror circuit amplify the voltage or current or both across the capacitor (470uf) with mirror circuit connected to capactor.
Another idea came to my attention, is that you are using the LEDs as outputs loads.I would expect those loads would recieve digital signals,high or low,(1 or 0 ), instead of a ramp type of voltage.

18. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
A current mirror repeats the current in the 2nd transistor as in the 1st transistor. It does this by creating the same voltage on the BE junction on both transistors. While not often used, there is a model for BJTs that suggest they can be treated as voltage controlled devices. The current model is much easier to use over all, and much more predictable, but both have their points.

It is important to note that the transistors much be matched (of the same part number) and at the same temperature. For this reason a current mirror is usually used as part of a semiconductor, an IC. Any variation between the transistor in both temperature and characteristics causes drift, sometime major.

It is a constant current sink in the above mode.

You can create a current mirror where the two transistors have different currents.

http://forum.allaboutcircuits.com/album.php?albumid=152&pictureid=1667

In this design, currents are a ratio between R1 and R2 or R3. If R2 is larger the current through Q2 is smaller, if R2 is smaller the current through Q2 increases.

Last edited: Aug 19, 2012
19. ### Thevenin's Planet Thread Starter Active Member

Nov 14, 2008
183
1
I use the RC in the e-book to get a better understanding about the circuit: F= 1/0.7RC=1/0.7*450 uf*100k = 0.0303951 . What is this answer referring to ? Yes I know it suppose to be Frequency.This is less then one.

20. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
There is nothing wrong with that answer. One cycle every 33 seconds (the inverse, or period).

But you got your math wrong, the answer is 0.317 Hz. In that case it would be 1 cycle every 3 seconds (approximately).

F = 1 / (0.7 X 450e-6 X 100e3)

Thevenin's Planet likes this.