Operate delay relay

[ptonsparky]

Small timer that needs to delay for one second upon being energized. Timer is 1-180 (all we had in town) seconds. Accuracy is 5%+-, repeatability 2%+-, recycle is 100ms.

Is Acc 5%+- of setting? With the repeated time being accurate within 2%. The power must be removed for at least 100ms before timing again.

So at 1 sec I can expect .95 to 1.05 and the repeated time being within the 2% of that for a total possible variation of .14 second. I'm confused so hopefully someone can clear this up.

 

[Ingenieur]

The mfg/pn and data sheet would help
it must be of setting, not full scale
5% of fs is 9 sec, pointless to have a <9 sec setting

 

[ptonsparky]

This maybe if it came. Time Mark signaline

 

[GoldDigger]

Small timer that needs to delay for one second upon being energized. Timer is 1-180 (all we had in town) seconds. Accuracy is 5%+-, repeatability 2%+-, recycle is 100ms.

Is Acc 5%+- of setting? With the repeated time being accurate within 2%. The power must be removed for at least 100ms before timing again.

So at 1 sec I can expect .95 to 1.05 and the repeated time being within the 2% of that for a total possible variation of .14 second. I'm confused so hopefully someone can clear this up.

The time will always be between .95 and 1.05.
If it happens to be 1.01 the first time it will fall within a 4% band that includes 1.01 after that. It could have a range of .99 to 1.03 or it could be 1.00 to 1.04.
If it hits .95 the first time, it will be between .95 and .99 after that.

 

[Ingenieur]

http://time-mark.com/Documents/Datasheets/Model 330_360.pdf
doesn't spec of setting or of range
must be of setting
pointless to have a 1 sec setting if error is +/- 9 sec lol

 

[GoldDigger]

I would ask what the note attached to the "**" says.

 

[Ingenieur]

I would ask what the note attached to the "**" says.

At rated/ambient temp on his sheet
specs the base pn on mine

 

[ptonsparky]

http://time-mark.com/Documents/Datasheets/Model 330_360.pdf
doesn't spec of setting or of range
must be of setting
pointless to have a 1 sec setting if error is +/- 9 sec lol

Yea, I kind of figured that out.:slaphead:

My circuit wasn't working ealier today for what I wanted. Simple deal, I checked and rechecked all the wires, made sure my bifocals and OFitis wasn't the problem. In a way it was, my help had brought me an interval timer and I didn't notice for some time.

The delay timer works but i may have to add a bit more delay cause it quit doing its thing 10 minutes after I left. That .04 sec GD explained doesn't sound like much but in this case it very well could be.

 

[ptonsparky]

I would ask what the note attached to the "**" says.

I could not find the note.

 

[Electric-Light]

Small timer that needs to delay for one second upon being energized. Timer is 1-180 (all we had in town) seconds. Accuracy is 5%+-, repeatability 2%+-, recycle is 100ms.

Is Acc 5%+- of setting? With the repeated time being accurate within 2%. The power must be removed for at least 100ms before timing again.

So at 1 sec I can expect .95 to 1.05 and the repeated time being within the 2% of that for a total possible variation of .14 second. I'm confused so hopefully someone can clear this up.

What is this for? Don't count on far ends being dead-on.

I am not sure what accuracy means, because the dial setting does not give you a direct input. It is like tuning an old school radio. Repeatability how consistent it is over multiple cycles.

 

[GoldDigger]

What is this for? Don't count on far ends being dead-on.

I am not sure what accuracy means, because the dial setting does not give you a direct input. It is like tuning an old school radio. Repeatability how consistent it is over multiple cycles.

And also at multiple ambient temperatures, for some designs.

 

[GoldDigger]

Yea, I kind of figured that out.:slaphead:

My circuit wasn't working ealier today for what I wanted. Simple deal, I checked and rechecked all the wires, made sure my bifocals and OFitis wasn't the problem. In a way it was, my help had brought me an interval timer and I didn't notice for some time.

The delay timer works but i may have to add a bit more delay cause it quit doing its thing 10 minutes after I left. That .04 sec GD explained doesn't sound like much but in this case it very well could be.

If you want a minimum delay of 1 sec, I would set it so that you are initially measuring 1.4. Hope that too long a delay does not cause problems too.

 

[Jraef]

What the "**" is saying is that the time values (where the ** follows the value) is valid at the 70F value that follows the ** further down. Note that the operating temperature range is 14 to 140F, but the accuracy ratings were taken at 70F. Bottom line, this is an analog electronic timer and the electronics are subject to changing their operations at different temperatures. We have been getting kind of spoiled with never digital products that can more easily have temperature compensation algorithms built in, but back when these were designed, that was difficult to do. The thing is though, despite that issue of drifting accuracy, these older circuits were often much more reliable.

 

[gar]

170308-2426 EST

Suppose this timer or time delay is an analog device, then it is probably based on an RC timing circuit, a threshold means, a charging source, and associated gates, switches, and memory means.

To design a stable timer the charging circuit and the threshold detector have to be individually stable or have to track each other.

Highly accurate variable resistors are not cheap. Thus, a major factor in accuracy is the adjustable resistor. Accurate fixed resistors are quite inexpensive. Accurate capacitors are not cheap. The fixed resistor determines the minimum time in combination with the capacitor and threshold. The fixed resistor plus the maximum value of the variable resistor in combination with the capacitor and threshold determines maximum time.

Probably to build the most accurate timer I would select (match) the fixed resistor to the individual value of the variable resistor. This will do nothing for accuracy between min and max time. That is dependent upon the linearity of the variable resistor. i would use a moderately accurate capacitor (5 or 10% tolerance) and then make the threshold an adjustment to set the timer for a specific time at some point in its range. For example the minimum time.

So I have no idea what the accuracy specification for this particular timer relates to. Repeatability has to do with noise, temperature, supply voltage, mechanical stability, and other factors. So set timer to its minimum value and measure the duration. Do the same at maximum time. And in between if you want.

Adjustment for 5% of setting is probably easy at minimum, use a 1% resistor, but now I still need a fairly accurate variable resistor.

.

 

[gar]

170309-1455 EST

If you go to the Potter & Brumfield Datasheet for a CLF-41-70010 Time delay to operate there is a timing specification that makes sense for an analog RC time day device. See http://datasheet.octopart.com/CLF-41-70010-P-datasheet-20740057.pdf

Expected range "Knob Adj. Types:–0, +20% of max. specified at high end of timing range; min. specified, or less, at low end."
What this means is that components used are run of the mill tolerance.

For "Fixed Types: ±5%". This means you have to adjust your resistance from a calculated value to get a more exact time value.

" (for AC units add ±1/2 cycle 60 Hz.)".

On repeatability " (for AC units add ±1 cycle 60 Hz.): ±3%" not clear if this is meant to include temperature and voltage.

The percent values will be of setting.

Back in 1975 when using some of these relays in automotive differemtial assembly machines where the electrical cabinet was directly attached to the back of the machine we had relay failures about every 3 months. Internally the relays were built with capacitors where both leads were at one end. Because the capacitor is a mass cantilever supported from the two small leads there was a mechanical oscillation that occurred every machine cycle from machine mechanical shock and in a moderately short time the leads had fatigue failure and the capacitor broke away from its PC board.

The solution: open the new relay and epoxy the capacitor in place.

.