Will this timer work?

Hey Eric:

Without seeing the circuit I can't say, but to just do what you want a 556 should be adequate (at least if you can trigger one section with the other

-- I can't remember offhand if the 555 polarities work out).

The four possibilities that I can see for your board being different are: (A) the original designer just didn't like the 555, had his head up his ass, or both; (2) the 555 causes some problem that neither of us is foreseeing; (third) For some reason the 555 solution was more expensive at the time of manufacture; (finally) the board does more than just controlling stitching, perhaps being used with more buttons and knobs in a fancier accessory controller.

I presume that you would prefer to have a stitch period knob and a stitch duration knob rather than a stitch-on duration knob and a stitch-off duration knob. The former is easier with a 556, the latter can be done with a 555.

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

Greetings Tim, I want the stitch on and stitch off situation. I thought that a 556 would be needed for this. It seems the circuits I have made with a 555 wouldn't let me do this. I guess I need to look some more. Thanks, Eric

Regarding (2), I have never been a big fan of 555s due to the huge current spikes they introduced on the rails. (Not to mention the plethora of supporting parts required for even the simplest timing jobs.) I have read (somewhere) that the spike problem has been solved (maybe only in the CMOS versions?), but maybe that was what the original designer was concerned about.

Best regards,

Bob Masta DAQARTA v7.60 Data AcQuisition And Real-Time Analysis

Scope, Spectrum, Spectrogram, Sound Level Meter Frequency Counter, Pitch Track, Pitch-to-MIDI FREE Signal Generator, DaqMusiq generator Science with your sound card!

I just think of a 556 as two 555's, so I tend to use "555" where I mean "two 555's".

If you want to control stitch-on and stitch-off duration, then you bypass one of the resistors in a 555 astable circuit (I can't remember pin numbers off the top of my head, but it's the one that goes between the discharge pin and the cap, and you want the diode cathode toward the cap. IIRC. YMMV. AODA (all other disclaimers apply)).

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

--
Trying to do the whole thing with a single 555 is pretty nasty since 
even with a variable duty cycle used to control the weld time, as 
far as I know there's no easy way to change the frequency without 
causing interaction with the pulse width.  

Using a pair of 555s or a 556, both in monostable mode, would be 
appropriate, since you'd use one to determine the time between 
stitch welds and the other to determine the weld time. 

In use, what you'd do would be to start one of the one-shots when 
the trigger was pulled, and then, when that one-shot timed out it'd 
trigger the other one which, at the end of its cycle, would start 
the cycle anew, generating another round, forever, until the trigger 
was released. 

Would you like a schematic? 

John Fields

If you wanted separate frequency and duration controls, I'd implement it with one working as an astable, triggering the other as a one shot.

That does exactly what you say you don't want -- it makes the pulse width settings interact with frequency.

But somewhere in this thread Eric says that he doesn't want to control frequency and duration -- rather, he wants to control the duration of the on and off times individually.

In that case, a single 555 would work mighty well, and has the advantage that in the event of some oddball occurrence the system wouldn't stall out, with each timer waiting for a trigger from the other. Given that you've got a circuit with mA currents at 5V sitting next to one with 60A at tens of volts, I think that "oddball occurrences" should be planned for.

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

--
How would you do it with a single 555?

Look in the data sheet for the usual astable circuit. Replace each resistor with a pot + series resistor (and short, shielded leads given the environment!). But I'm not done.

If left at this point, then the pot from the discharge pin to the cap will have current flowing through it in both directions, and will affect both the on and off times. So, bypass it with a Schottky diode going from the discharge pin to the cap -- now each pot will independently adjust it's own time.

Ta da.

I can't remember where I learned the diode trick, but it was decades ago

-- I think I was still in school, and may have picked it up from one of my profs at Portland State.

--
www.wescottdesign.com

use diodes to steer the current.

----+--- vcc (+12V) +---v | off time +-[VR1]--->|--|--------+ | | | | | | +---v | | on tine +-[VR2]---| out1 (totem pole) | . LM555 . +-------TH(6) DIS(7)-----> out2 (open collector) | . . +-------TR(2) CV(5)-- C1 | . . ===== . GND(1) . | . . . .|. . . . | | +-------------+ | ---+-- gnd

VR1 sets the "output low" time VR2 sets the "output high" time If you use a bipolar 555 (NE555, LM555) the discharge pin can sink 200mA which at 12V is enough to drive a mid-sizd relay.

--

Bye. 
   Jasen

Greetings John, I think you have hit on exactly what I want to do. Yes, please, a schematic would be great. Thanks, Eric

Greetings Tim, I must be confused or confusing or both. What I want to do is control each function independent of the other. So if I increase the weld pulse time I don't want the non-weld time to change. And the other way around. So let's say I'm welding a 1 inch long bead and then leave a 1 inch space, and then weld a 1 inch bead and so on until I let go of the trigger. But then I decide I want 1/2 inch long beads. I change the welding pulse so that it is only half as long as before but the non-weld time stays the same. Isn't that what John Fields is describing? Or am I just confused and confusing everbody else with illiterate posts? Thanks, Eric

John is describing (I think! Maybe I'm the one reading it wrong!) a system where the start of each bead is a constant distance from the start of the last, and you can adjust the length of the bead.

I.e., you set the cycle length to two inches, then choose a weld length within that.

I could see wanting to do it either way, and possibly giving the user a switch to do so -- this starts getting difficult with 555 timers, which may be why you have something entirely different.

At any rate, both the circuit that I describe and the one that Jason Betts posted will work -- I think Jason's is a bit better than mine, frankly.

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

If you use a bipolar 555 you can't be sure the output goes high enough.

If you use the discharge pin straight to a relay, then even with a diode across the coil (which slows down the relay opening) you'll be abusing the

555's protection diodes.

I only use this arrangement with a CMOS 555, and I'd drive a relay with a separate transistor, using a resistor in series with a diode, or a zener diode, to make sure that there's a hefty voltage across the coil on turn- off, to speed up the current decay and make the relay turn-off quicker and more positive.

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

Greetings John, I though I posted a reply to your message above but when looking for it I don't see it. So, yes please I would love a schematic since it looks from your post that you are describing what I want. I think. Thanks, Eric

it seems reasonable to expect 11V with a 12V VCC. Why do you think that's not enough?

there's no protection diode on the discharge pin (what would be the point).

but yeah you need parallel diode on the relay. the relay's internal resistance will turn it off fast enough, I doubt we need to be faster than 100ms in this application. but if speed is needed a parallel RC in series with the diode can expedite turn-off withput incrasing the stand-off voltage.

--
umop apisdn

--
Hi Eric, 

I'm off to a doctor's appointment for a while, but I'll post the 
schematic and an LTspice simulation for you some time today. 

John

--
Here's a link  

https://www.dropbox.com/sh/yikxzbjuncr1v6l/AACoSK5d_R2YsKxZTIPGBY4Oa?dl=0 

to a schematic and to the .asc file that'll let you run an LTspice 
simulation, if you care to. 

I opted for a single 555 running as an astable because Tim Wescott 
was right; my original dual one-shot circuit could have entered a 
hang state with no output without a manual re-start, while this one 
is inherently unstable and _MUST_ run, barring some major disaster. 

One caveat is that its first output pulse will be about twice as 
long as the others, and another is that there's a very slight amount 
of interaction between the times when they're being set. 

That is, if you set the weld ON time for, say, 100ms and then the 
OFF time for 1 second, the ON time will change a tiny bit. Less than 
1% as I recall, but you can see it for yourself if you run the sim.  

If you decide to use the circuit, make sure that C2 is wired right 
at the package, directly across pins 1 and 8.  

John Fields

What would the purpose of that 1000 ohm resistor (R1)be? It is in series with a 1 meg ohm resistor so would seem to be unnecessary?

Owen

--- When U1-7 (DIS-) goes low in order to discharge the timing cap, C1, it'll also sink the current from the 12 volt supply through R1 and R2.

As a consequence, if R1 wasn't there and R2 somehow wound up at zero ohms, the 12 volt supply would be connected directly across U1-7 and GND.

Depending on whether it's a bipolar or CMOS chip, U1-7 will be internally connected to either the open collector of an NPN or the open drain of an N-channel MOSFET, resulting in the supply positive being connected to ground with only Rce or Rds limiting the current through the device.

If you run the sim with R1 and R2 shorted you'll see that, for the NE555 with U1-7 low, current into U1-7 will limit at 2 amperes, which means that Rce(sat) is running about 6 ohms and the package will be dissipating 24 watts.

Not a Good Thing, and the magic smoke will probably escape.

John Fields, Professional Circuit Designer.