Maker Pro
Maker Pro

Replacement Optical Isolator

Leggyownz

Feb 7, 2017
55
Joined
Feb 7, 2017
Messages
55
Trying to find a suitable replacement for an optical isolator. The datasheet of the one I have currently is below. The manufacturer of the board I'm working on upgraded these over 20 years ago and the upgrade kits are long gone. Trying now to upgrade mine but haven't found a replacement that fits the criteria in the datasheet. Also stuck between translation of the old terms in the datasheet to newer terms. Anyone know of anything that would work?

https://4donline.ihs.com/images/Vip...7-1.pdf?hkey=EF798316E3902B6ED9A73243A3159BB0
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
I doubt this type of optocoupler is made any longer. Today's optocouplers use LEDs and phototransistors.
Imho you have three options:
  1. find an original coupler (or similar one) on a flea market, ebay or similar (as stated here).
  2. build your own coupler using a neon lamp and an LDR coupled and sealed from ambient light by e.g. shrink tube.
  3. replace the coupler by a modern one. This will surely require modification of the original circuit to accomodate the very different parameters of a modern optocoupler.
Options 1 or 2 probably mean the least effort to you.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Or open up the existing one and replace the bulb (the most likely failure). If this operates from DC, you might get away with a white led and a suitable series resistor (that you might have to determine by trial and error).

Make sure you seal it up if you take this approach. You don't want stay light getting in.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
plus probably a diode antiparallel to the LED as the neon lamp is probably AC powered?

Now I'm not looking at the datasheet on a tiny screen I can see it's a neon. Aaagh!

How often do neon bulbs fail over time? -- if it was a filament lamp, then I'd bet the failure was there.
 

Leggyownz

Feb 7, 2017
55
Joined
Feb 7, 2017
Messages
55
Well the guy I know that understands these machines fairly well says the isolators get weak over time. Didn't specify too much further on that. Although we are talking about a board that's like 20-30 years old at this point, it's not a surprise. Well, I was hoping for easier answers but I should've known better on that. If I even had a datasheet, I might attempt replacing with something modern. Or if I was working on this at home for myself and not at work.

Although, the guy I've been talking to, was expecting incandescent, and not neon. Is it possible, it might be easier to swap from neon to incandescent instead of hopping straight to LED?
 

Leggyownz

Feb 7, 2017
55
Joined
Feb 7, 2017
Messages
55
I've done the ebay search before, 0 results and looked through the similar results but I still don't have the knowledge to tell whether any of the similar models would be interchangeable or what I would need to do to make them interchangeable.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Well, I was hoping for easier answers but I should've known better on that.

It would be really great if you can find a source for this part, but at least it's something that you can cobble together a replacement. There's lots of stuff that while once common is now extreme unobtainium.

Although, the guy I've been talking to, was expecting incandescent, and not neon. Is it possible, it might be easier to swap from neon to incandescent instead of hopping straight to LED?

The problem is that the drive requirements are so different. If we had access to the circuit diagram then we could probably design something to use a more modern replacement. Without that, you need a solution that will work with what you have. If it had been an incandescent bulb, a LED may have been an option due to its lower power requirement. However neon bulbs are a high voltage low current device that is very different in characteristics to an incandescent bulb or LED.

Assuming the vactrol isn't working... My approach would be to note the exact placement in the circuit before removing the device. Then I would use a multimeter to confirm that one end is open (the neon) and the other has some large value of resistance (the LDR).

If I could confirm this, I would carefully try to get my way into the sealed enclosure, working from the end containing the neon tube. Ideally I would find that I was able to separate both halves in a way that I could push it back together again.

The next thing to check would be that the LDR end changes resistance markedly in response to light. If this is the case then we can assume the neon tube has failed.

I would then go looking for a similar neon tube and try to replace it. After squeezing stuff back together as best I could, I would use something light-tight (NOT insulation tape) to make the enclosure light tight again.

HOWEVER, it is possible that the problem is not in the vactrol -- are you sure that the rest of the circuit is operating correctly? If you've tested this, great!
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
In the original circuit, is the opto-isolator being used simply for switching something on/off, or is the CdS cell at its output being used as a variable resistor for some analogue function?
 

73's de Edd

Aug 21, 2015
3,613
Joined
Aug 21, 2015
Messages
3,613
Sir Leggyownz . . . . .

In my hand, I am now holding, one each quantity, of a . . . . . virgin / new . . . . . . .VACTROL VTL3B49 electro optical isolator.
Of which, I just cut in half , lengthwise, to inspect. . . . . using a micro cutoff wheel in a Dremel hand grinder tool.
The cross slice was being made 1/16 inch down from its 8218 * code date marking and another at the 180 degree opposite side..
( * The 18th week of 1982. )
The case is being of black polycarbonate and you can judge its thickness by a side view and then proceed VEWY- VEWY slowly for the very final increments.
The lamp end has a NE-2 (H) lamp with super full length 1 1/2 inch leads, as does the other sides serpentine routed 13 row photosensitive strip detector. The tip of the lamp seems to almost touch the face of the sensor.
Additionally, the internal surface of the black poly tube has been spray painted with white lacquer for maximum reflectivity of the provided neon illumination.

Now you certainly did not give me any information on WHAT this piece of equipment this is within . So I know not whether this is being a control board for an industrial sewing machine . . . . or a separator for bull semen extraction . . . please be more forthcoming with further more definitive information .

The only smidgen of info was the . . .

Well the guy I know that understands these machines fairly well says the isolators get weak over time. *****

There are two failure aspects . . . . the photocell and the lamp.

Initially in considering the photocell, I also have some slightly larger units that use a 1/2 in diameter photocell case /housing with a sealed glass insert at the end. If you are familiar with the quality look of a PIR sensor , that is what it resembles . It uses the same NE-2 lamp facing onto the photocell.s sensor..
On those premium units, on the sensor unit itself, I would expect a half life failure rate of a 100 years.

On the lamp, I would expect that to depend upon HOW much use time it is activated.

Specifically . . . . of that era . . . we should fully expect the control electronics to be solid state . . .but I feel that this unit is associated with a go no go or fail safe condition, such that an 120.240 AC is present at the lamp and the loss of that voltage then causes the sensor portion to interface with the solid state aspect of the control board and indicate/react to that condition.

Now if this application is being correct . . . . . that will then tie in with the ***** statement above.
Since, with the specified current limiting resistor used with the neon lamp, expect the lamp to incrementally start developing internal ion burns with extended use and start a darkening of the inside of the neon lamps glass .
That then accounts for a tapering off of the light output intensity available to reach the photocell.

upload_2019-4-6_4-54-50.png


ASIDE:

RE. . . . . .How often do neon bulbs fail over time?

I see it all the " maternally copulating " time . . . on power strips with internal neon lamps as power on indicators.
Either they are OUT or are intermittently flickering.
HINT . . . far east lamp manufacturers . . . you have already missed the products hey day for the last 40 years, but seems like you FINALLY should have realized the fact that you need a slight trace of radioactive elementality for ionization enhancement.
Such as has been constantly used by all the reliable neon lamps made by Gentile Electricity and Chicago Miniaturizing lamps.

Thaaassssssit . . . .

73's de Edd . . . . .


All things being equal . . . . . fat people use more soap.

 
Last edited:

Leggyownz

Feb 7, 2017
55
Joined
Feb 7, 2017
Messages
55
Currently checking to see if my guy has circuit diagrams for this. It's a floor centrifuge, Specifically Beckman J2-21.

The neon optoisolator acts as a switching device and this uses electromechanical relays, not SSR's.

I may have missed some questions or some information that would be helpful, if I did just point it back to me again. I'm absorbing a lot of information over here. I'll automatically reply again when I get word back on whether I'll be given a circuit diagram or not.
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
An H11F1M optoisolator (one of these), with a suitable input current-limiting resistor might do, providing the output terminals don't see more than 30V across them.
 

73's de Edd

Aug 21, 2015
3,613
Joined
Aug 21, 2015
Messages
3,613
Sir Leggyownz . . . . .

Wheeeeeeee . . .now we have eliminated multi-millions of possibilities . . . . of what the product involved actually is.
In reality, a "blud slanger"

And is the supplied photo of the power board, in actual agreement with being like your unit ?
Proceeding, in that consideration, I am seing the power section of the system with two other boards handling the control and logic aspects.
On the power board shown, I have marked up the TWO involved neon based optical isolators as the two YELLOW A units.
And considering these to be factory unirs, with the 211-427 ? OI units seems that these are using the larger 1/2 in dia photoresistance detector unit instead of the 1/4 unit used in the VACTROL VTL3B49 unit that you / I have.
With due diligence that certainly the two are interchangeable, with your unit working satisfactorily in past use with the VACTROL subs of the mods..
My big ones were of FAIRCHILD INSTRUMENTS manufacture.
With further full use of my 18" crystal ball and thorough observations around the board . . . within detail limitations .

WHAT I SEEM TO SEE . . . . .

The far right RED outlined BLUE E-cap and its White Red Trace lead over to its far left is its RED B block FWB rectifier companion, thereby seeming to relate to being a 24VDC power supply, with possible further supply voltage subdivision reduction for the 2 other logic level boards.
Back to B FWB and at its forefront is a SOLID STATE relay at X. On the top side direction of B FWB, seems to be one BRUTE of a 3 terminal power relay (220VAC). With it having had a later add on protective MOV block at the RED STAR.
At center panel foreground we have a 3 pole open frame power relay at Green outlined D.
Suggestive of being 220VAC power function switching with a 24 VDC coil.
So o o o o o, now we have 3 power control relays, and this unit is needing to handle the HEFTY power demands of a spin motor and a refrigeration cooling pack.

Now the minor aspects . . . the two YELLOW outlined neon based Optical Isolators at the two A positions.
The left ones close proximity to the D relay suggests of it wanting to be relating to AC voltage sampling of that units condition. Thus the reason of the simplicity of a neon lamp isolator, as it only requires a droping /current limiting resistor for interfacing / sampling of supply AC power voltage..

Looking over to the right A, is the like designated Optical Isolator, now it could be arranged as several connections, with YOU having to peek at the bottom side trace foils for circuitry routing.

Now, it could be that if that F designated carbon resistor is in the 20's - 30's K's of resistance, the White / Red (or violet ?)trace wires just above it go to the monitored AC voltage of interest, and then it series feed thru that resistor to that lamp end of the OI.
That then leaves the left sensor ends 1 lead to go to ground and the other lead to that E capacitors right lead, that otherwise seems to go to nowhere. But would relate to receiving logic input from the sensor.
YOU will know immediately, since you can read the lamp-cell ends markings on your OI. I don't have that advantage.

Now, the opposite wiring consideration would be that double RED marked set of diodes are configured as a discrete FWB set, as is being fed from AC at the ORANGE and White / Blue trace wire sets just above them.
The positive DC output passes down to the . . . probably .1 ufd E designated cap. (Since the larger companion just below is a .22ufd.) That DC voltage then feeds the Neon within the left end of the OI, and has the other ends photo sensor feeding out to the F resistor and the White / Red (or violet ?) trace wires, carrying logic output.

Another unknown , until you confirm the situation, seems like a YELLOW Sprague, little Atom, E-cap is the other double RED boxed cap . I suspicion it to be a minor supply storage E-cap that might be having its rectifier set being the 4 discrete diodes constitutig a FWB that are hidden by the wiring cluste down near the caps negative lead. ( Since its Positive leads shell clamp ring crimp shows at the top .)
Passus the info on that YELLOW units capacitance and voltage rating.

ANALYSIS . . . . .
If you have to work naked, without a schematic, you need to make voltage checks at the lamp terminals to see if the lamp is receiving DC or AC and I highly suspect AC . . . . . by virtue of my suspicion of the OI's actual functionalities.
DCuse would only have one electrode glowing and a definite decrease in light output.
It takes up in the low-mid 70 volts to flashover and fire the neon lamp and then it reads in the 50-60 volts while running thereafterwards.
On the sensor end you would now want to take a like reading across its terminals for both DC and AC .
I am suspecting DC, with little or no AC presence. Then you take note of that DC voltage in a system operating state, when the neon lamp is activated.
Then shunt a 47K resistor across the OI lamp leads, right AT its case and take a second reading of the DC voltage, then the effective resistance of the sensor at a nomal state of illumination can be cross computed.
Then you would need the same new OI unit which I have 1 of, for you , to then sub in place of your unit, and establish it as being a benchmark standard , to see if your OI units neon lamp has received enough usage, so as to have started a cumulative and progressive greying of the intenal glass and an accompanied declining light output illumination to the sensor.

Thaaaaaasssit . . . . . . . . . .


PHOTO REFERENCING . . . .

(https://i.ibb.co/k6nKCY9/Beckman-J2-21-Centrifuge.png)

Beckman-J2-21-Centrifuge.png


73's de Edd . . . . .

It's not IF or HOW you pick your nose . . . but . . . WHERE you put the booger.

 
Last edited:
Top