Combining breakout Amplifier and Opto-Isolator

41

Hi John

I find if there are a few messages on the Email bit it is sometimes confusing getting the chronological order right so I always use the forum to answer. Seems easier to me anyway.

I can see that now and I can see what you have done. That is what I asked, to advise exactly what IC has been used.

You originally said you used the BOB09816. At no time have you advised you only used the circuit and produced your own board. I got worried when I discovered the Sparkfun board had used a 5V IC and I had advised you to use 12V. When you previously said you had the higher voltage IC I thought Sparkfun had made an error. Then yesterday when I looked closer at your sketch when I discovered I had guessed the wrong MP3 trigger board I saw the OP amp board was different too. I must have spent 2 or 3 hours trolling the web looking for a board that looked like the pic. Then I cropped it and blew it up it started to look a bit home made. At this stage I couldn’t read any resistor values as the image was too pixelated. Then you come up with the latest little gem confirming the board is indeed home made using the higher voltage IC.

The point I am making here is that it would have been a great help if this had been made known right from the word go. Firstly you still have not confirmed the part no for the MP3 trigger board. I guessed the wrong one and did not realise until I saw picture of your set up. Secondly you could have saved me a lot of worry and searching if you had said up front you had made a larger board and used a different IC for the OP amp breakout. I knew you wanted a better volume control but did not know you had built a complete board. While we are at it have you got any more little bits of information you think may be useful.

You need to remember I am trying to help remotely and am not looking over your shoulder to see what you have done. When you are describing things accuracy is paramount. Although I am retired I don’t have all day to sit in front of a computer guessing and hoping to be nearly right.

Cleared up a couple of issues, yes.
Cheers Bob

1 Like
42

I understand your frustration, and I would be frustrated too. I can see where you are coming from and I apologize for being the cause of 95% of your frustration.

Some of the searching etc was due to an assumption on your part. You advised to use 12 V. And I confirmed I did.
I took that as OK and moved on towards finding the overall solution.

And I accept that here is one of those instances of me being focused on what I was trying to achieve, I confirmed the 12V operation and looking further ahead to find the overall solution and you were not here looking over my shoulder to see what I had done with the op amp board.
As far as I was concerned, I had confirmed the voltage and apart from that, the rest of the circuit is exactly the same as the Op amp breakout.

Where you say that “you still have not confirmed the part no for the MP3 trigger board.” 
 as far as I can see there is only one. And therefore I thought you knew the board I was working with. The only part numbers I remember you asking for was for the LM358. Sorry if I missed your request for the MP3 Trigger number.

I did not know about the “Qwiic MP3 Trigger” and now that you have pointed it out I do not see that as being the same. But once again, in my mind, and, with you not being here to look over my shoulder, you did not know or see that I had only ever focused on “contact closure triggering”. Hence I would not have even been considering the Qwiic MP3 Trigger.
So I need to be extra vigilant on confirming details for you.

So, following your recommendations and with some testing different resistors I have now settled on using 120 ohm resistors on the output of the transformer and 51K resistor to replace the 100K resistor as you suggested.

I am now working on switching the output of the MP3 Trigger to feed into one of two applications

  1. Thru what we have been working on
  2. Thru a 0.5W amp into a 0.5W speaker so the MP3 files can be activated locally, without broadcasting them onto the system. Checking and testing.

Regards
John

1 Like
43

Hi John

Due to the lack on accurate information initially it had to be ALL assumptions.
This has gradually declined as information gradually trickled through.

Rubbish. You have used a different IC to the one on the board. That bit had me worried when I realised it is only 5V and I had advised using 12V.

And you STILL have not confirmed which board you are using. I have only guessed again by comparing the photos.

More rubbish. Core web site list two. DEV-16892 which is the one I thought you were using and WIG-13720 which is the one I now believe you are using. The DEV-16892 has an on board speaker amplifier hence my earlier reference to this output.

I don’t know what Qwiic is either so had not considered these anyway.

I don’t know or care how the triggering is done. I had always thought you know what you want here and have this sorted.

The resistors in the transformer secondary circuit were your suggestion and I thought a good idea as the 33Ω (worst case scenario) load was far too low for that 358 IC and their inclusion would lift the load to a better value and help any overcurrent clipping problem. This has presented another problem. Signal level. If we say the max P-P voltage with 12V supply is 10V. Disregarding transformer losses you should have 10V P-P across the secondary. The two 120Ω resistors and worst case 33Ω load will form a voltage divider resulting in 1.2V P-P across the load. This equates to 0.427 (427mV) RMS. No where near the 1.5V required.
This is getting to be a circular problem and getting back to my original questioning of the ability of the 358 amplifier to drive a low impedance load. This amp normally has a load up into several ks and the data sheet is quoted using loads of 2k and 10k.
I suggest you carry out your experiments but I think I am right. You may have to go back to my earlier concerns re the LM358 and rethink about using an IC that will drive loads as low as the 33Ω (worst case).
You would then be able to use the 600/600 or another suitable transformer without the extra resistors and get your required voltage output. As most of these circuits drive a speaker direct via a capacitor and one side of the speaker grounded simply replace the speaker with your transformer.

Regarding your audio input from the MP3 trigger. It is easy to have both L and R channels combined, especially with your home-grown board. If you duplicate C5 and R2 and take them to the same point (IC pin 2) the OP amp will combine the stereo channels into one mono channel.
Cheers Bob

1 Like
44

OK thanks Bob,

MP3 Trigger - SKU: WIG-13720 is the only one being used.

Re “Qwiic” version
I did not know about the Qwiic version and you don’t know what Qwiic is either
This is the one you found - SparkFun Qwiic MP3 Trigger - SKU: DEV-16892
This is not being used.

Both of the above highlight my mistake of listing names instead of model or part numbers. Sorry.

The info in the last reply about not enough voltage to get to 1.5Vrms and the effects of low impedance
and you have mentioned an IC that will drive loads as low as the 33Ω
Can you clarify something for me please ?

I have looked back through the posts, looking for a chip number for driving, say, 33 ohms. But I can’t find anything other than referring to a class D amplifier.
Did you mention a replacement for the LM358 that would drive low impedance load ?
Or do I just look for a class D speaker amplifier ?

Can you recommend one for me to try ?

Regards
John

1 Like
45

Hi John

That referral to a class D amp was the one on the other MP3 trigger board (DEV-16892). I would stay away from simple Class D amps a bit as most of them rely on the speaker inductance to filter and recover the audio and really become part of the output circuit. I only mentioned it as something you might try. Your load is likely to vary a bit and even though we are looking at a transformer if the load is resistive (33Ω resistor + 100Ω pot) the reflected impedance will be resistive and actually look like a resistor, not an inductor. Strange I know but true. I think some class D amps include inductors in the output circuitry to provide a known inductance and the value has to be changed to suit speaker Z.

I had in mind the LM 386. Have a look at the data sheet. It has a gain of 20 which can be increased to 200 by the addition of a couple of components. I am going to try to get to Jaycar one day early in the week so I will grab one and do a couple of experiments. This unit data sheet says max load to be 32Ω and with a 600/600 transformer the load would be anywhere between 33Ω and 133Ω you may have to think about your 10k / 2k one with the 2k side as primary. This will present a load between 6.6Ω and 26.6Ω which is in range. If your little transformer won’t handle the bit of extra current you may have to purchase one that will. Quoted impedance I don’t think will be totally important as long as the turns ratio is about 2:1 or an impedance ratio of 4:1. Element 14 have a couple, one is fairly inexpensive and is quoted 600Ω / 150Ω. That is the sort of ball park to be after. If it gets to this try your transformer first.
Cheers Bob

1 Like
46

OK thanks Bob,

I have had a look at the LM386 info. I imagine the gain of 20 will be sufficient. Is this correct ?

The 10k/2k transformers I have from Altronics are
a) in short supply and do not look like being replaced with anything
b) do not list any specs nor has datasheet
So I would prefer to go with something with better availability

Looking at the Element 14 transformer you mentioned, and your comment of “If your little transformer won’t handle the bit of extra current you may have to purchase one that will.”
I have three questions.

  1. Is the Element14 transformer you mention (600R/150R) model 104S, this one ? https://au.element14.com/hammond/104s/audio-transformer-0-006a-600-150/dp/2546919?st=600ω%20/%20150ω

  2. Is this “extra current” on the primary, or secondary, or both sides ?

  3. How do I work out what current handling capability I need ?

  4. The data on that transformer says Secondary Power - 500 mW. Will this rating be high enough ?

Regards
John

1 Like
47

Hi John

Should be. It would not hurt while experimenting to bring the gain pins 1, 8 to a terminal block or something in case it has to be increased. This will make changing components easier.

Good idea. Some suppliers can be very vague and short on pertinent details. That is one reason I look at Element 14 or RS for info. If there is a data sheet available they will publish it.

NO. I had a quick look and the low frequency performance is not that great, starts to roll off at 200Hz.
The one I looked at is Stock no 2280535 which is 600 / 600 or 150 depending on connections you would use 150. The other one is stock no 161410 which is 10k / 2k. Also much cheaper. At $8.01 each it probably would not hurt to get both as I have no idea at the moment what the projected performance will be. I still have some reservations about connecting more than one low impedance source to a balanced line at the same time. I have never done it as previously stated.
These transformers should be connected with the low impedance side as primary that is toward the amplifier and the high side as secondary connected to the balanced line.

Both

Ohms law

Should be plenty.

The last 2 queries have too many variables to give a definitive answer.

We have been discussing the LM358 and extra resistors in the transformer secondary circuit. Have you tried this yet or you leap frogging. If you don’t try things before discounting them you can very easily find yourself going in circles. If this looks like happening I am outta here.
Cheers Bob

1 Like
48

Thanks Bob,
If you see me starting to go round in circles, let me know.

I mentioned before that
“I have now settled on using 120 ohm resistors on the output of the transformer and 51K resistor to replace the 100K resistor as you suggested.”
But I guess I forgot to tell you that I settled on that because I had been changing resistors around and testing, and that combination seemed to work the best.

My concern is that you mentioned the resistors in the legs of the transformer would be reducing the voltage, and if I remember correctly, reducing by about 60%.
So with this in mind, plus you mentioned that the MP3 Trigger is looking for Kohm to feed into and not low ohms below 130 ohms, my thoughts are how long will this last with it not being set up to normal conditions. Also, I believe that the closer to normal 1.5 Vrms that I can get, the better it is likely to perform when it is hooked up to the real system and would probably handle any unknowns better than the other setup that is pushing boundaries.

Hence looking at something that drives low ohms and can be closer to the 1.5 Vrms.

John

1 Like
49

Hi John

Yes you apparently did.

Just what in the hell are you doing.
At no time did I mention the MP3 trigger working into low Z. My understanding is the MP3 trigger is connected to the LM358 board which has an input Z of 10k.
It is the LM358 working into very low Z that I was concerned about.

This is what I mean by going in circles. I am getting a bit sick of repeating. Are you perchance making any notes as you progress. If not WHY not. Unless you have an extremely good memory you will not have much chance recalling what happened with any particular situation.

If what you said in the par above is a mistake this is the sort of thing I mean. You need to get your act together and be very accurate or I will be off. I am flying blind enough without this sort of uncertainty.

As a matter of interest what do you have in the way of test and measuring stuff available to you. It is not much good asking you to measure anything if you are unable to do so.
Cheers Bob

1 Like
50

No. The “MP3 Trigger is looking for Kohm to feed into” should read LM358 is looking for Kohm to feed into.

Measuring equipment is
Digital Mutlimeter
I can get access to a digital CRO in short notice
Capacitance meter
Capacitor tester
In circuit transistor tester
Impedance meter - basically for 100V line speaker system
Insulation tester

1 Like
51

Bob, on a different topic and one one that you are uncertain about, I think I can clear up your concern about multiple low impedance devices on the “Line” at one time.

The following is how the system works all the time.

The following pics show an expanded view of the Talk line part of the circuit (Also known at the Party Line L1 and L2 on the drawing) for the Comms Stations
and a zoomed in part for what I believe you are calling the low Z part of the circuit

Any time a handset if off hook, relay K1 is active and the Comms Station is connected in parallel to any other comms stations that can be also off hook.

You can have 2, 3 or 4, maybe more all connected to the same line at the same time. Certainly the most common is two people talking, but there can often be three talking and there is nothing to stop others from picking up a handset and automatically be included in the conversation.

Is this the concern you mentioned about connecting more than one low Z device to the line at one time ?

Relay connect to Talk Line
Relay connect to Talk Line1459×360 118 KB

Low z connection
Low z connection1520×315 88.1 KB

Regards
John

1 Like
52

Back to transformers.
I could not find the 161410 which is 10k / 2k.
It took me a while to find, but I believe the number should 1610410. Is this the correct one ?

Following on from this, I have looked through the various options on Element 14, and because I don’t understand how to find the voltage and wattage required, and because I can get a 600 / 600 z transformer that can also be wired as 600 / 150 z, and, it has a power rating of 1 Watt, I am going to order one of these for testing. Element 14 Stock code 1610412.
1610412 600-600 1W transformer

Please let me know if I am wrong or off track with this.

Regards
John

p.s. I mentioned a CRO as test equipment. This is probably older terminology now for older technology and I believe the one I have access to is a relatively new one and would be digital.

1 Like
53

Hi John.
I will answer these 3 posts in order.

almost mandatory.

The only instrument to use if the waveform is anything but a sine wave. Essential for measuring complex waveforms such as voice or music. As I don’t see an audio generator listed I think you are using the music on the card. Correct?? Procedure is (if you don’t already know) measure the voltage P-P then divide by 2.83. This will give you the RMS value IF THE WAVEFORM HAD BEEN A SINE WAVE. A digital multimeter would be useless here unless it happens to read PEAK or P-P V (if PEAK divide by 1.414).

Handy if you have unmarked or hard to read caps.

Not a lot of use in this application.
Cheers Bob

1 Like
54

Hi John.
The second of three.
I think I am getting the hang of this system. I will outline a scenario which I think is correct and please correct me if I am wrong.

Joe wants to contact Bill to discuss a subject.
Joe picks up handset which will connect to Party Line.
Joe presses Page key which connects to page line and mutes local speaker, Joe pages Bill.
Bill picks up which connects to Party Line and conversation takes place over Party Line.
Joe or Bill decide they have to include Harry in this conversation. One of them pages Harry (Paging Line).
Harry picks up and joins 3 way conversation over Party Line.
None of this conversation is heard in speakers and will remain private unless some one else picks up a handset, in which they can join or listen in. Speakers in units involved in Party Line conversation are muted.
Your MP3 music is inserted on Page Line and will (hopefully) be heard in all speakers except the ones off hook.
The only time there is anything on the page line is when somebody actually pages on this line.
This may produce odd effects such as voice and music at the same time or it may be OK. The only way to find out will be by trial.
I think this correct.
Cheers Bob

2 Likes
55

Hi John.
Third of 3.

Yes. Sorry, typo on my part.

I only had a quick look and yes this is a better one than the one I quoted. Having both sides centre tapped allows a lot more flexibility and the 1W is good. Always better to go up if in doubt. You can’t get into too much strife that way. A bit less insertion loss also.
For your application you need the lower impedance on the primary side. Connect pin 1 to pin 3 and pin 2 to pin 4. Pin 1/3 to amplifier via capacitor and pin 2/4 to ground. Doesn’t matter which way around so can reverse if more convenient.
The secondary side will be the higher impedance so connect pin 6 to pin 7 and pins 5 and 8 to your paging line. Should be no need for any resistors or capacitors.
With a resistance ov between 33Ω and 133Ω ( I would expect this to be something less than 133Ω) this would reflect between about 8Ω and 33Ω in the primary side. Within recommended range of the LM386.

I think you are OK. Only a trial will tell. One thing to be careful of. If you are lashing this up on the bench do not apply any music to the 386 without a load of between 33Ω and 120Ω across the transformer secondary. I don’t know how this IC behaves with a high impedance load. Probably nothing but does not hurt to be careful.

Don’t worry, I have an old head (85) and recognise the term. As a matter of fact I used it a few months ago and I don’t think too many knew exactly what I was talking about. It probably is not the correct term these days as it stands for “cathode ray oscilloscope” and the “cathode ray” part is long gone.
Cheers Bob

1 Like
56

Hello Bob,
Re LM386, I have looked through my box of bits and I have one of the following LM386 based amplifiers.

I was using one or two of these in the past for a different purpose, but as far as I can see, it should be suitable for this current project, but I have a couple of questions that you may be able to help out with.

Firstly, you mentioned some amplifiers are more suited for use with speakers, rather than just audio like we want.
Is it the case that, as long as it is LM386 based with a few components like this one, that we will be OK using the one in the link above ?

Secondly, In the past the volume pot on these devices was a 100K pot. And this is shown in the picture. However, the one that I have now, has a 50K linear pot for the volume adjustment. I know this because I removed it from the board and the board is labelled 50K and the pot is also 50K linear.
I did notice that before removing the pot from the board, when I tried to measure across it with a multimeter, is was open circuit. But on checking, I could see that the trace from one of the outside pins on the pot, ran to a capacitor. So removing the pot from the board, and measuring, confirmed it is 50K linear. And yes I know that we should be looking at a log pot instead of linear.

I ran the signal from the MP3 Trigger into the amplifier in the above link and then out to a horn speaker and played an alarm tone. It seemed to have too much volume, and so I removed the 10 uF cap across pins 1 and 8 in the amplifier to bring the gain down from 200 to 20.
This sounded better, but still heaps of volume and I am not sure if the MP3 tone has some distortion in it, or whether it was the whole setup that was causing some distortion in a couple of places as the tone was playing.
I will have to get that oscilloscope and see what I can find.

In the meantime, every other circuit that I can find about LM368 amplifiers, uses a 10K pot for the volume control.
Are you able to tell me what difference it would make between using a 10K pot and a 50K pot for volume control ?

And do you think the device on the above link will be suitable for what I am trying to do ?

Regards
John

57

Hi John

True. The LM 358 for instance will not drive a speaker.

Not true. We (or you) are trying to drive a low impedance load. Not quite as low as a speaker but low enough. By the time you insert the transformer the load will be somewhere between 6Ω and 30Ω. The 6Ω is near enough to speaker. The 30Ω is a bit higher but all that means it will probably not produce as much power.
I am getting a bit sick of repeating all of this so will not be doing it again.

It would be nice to see a circuit but that unit should do the job.

What level were you feeding into it. Could be overload or it could be the quality of audio out of the horn speaker. It is probably not renowned for its HiFi qualities. Also what impedance is it.

Good idea. I wonder how you have come this far without it. As far as I am concerned it is the most useful instrument in the arsenal.

It might kill the signal level a little bit but to answer that I would have to know the impedance of the audio source. I would personally go for 10K but it should not matter much.

If you use a linear pot you will find that control down the bottom end will be very touchy and seem all scrunched up. It is all to do with how the human ear behaves. Electrically it makes no difference.

Are you going to answer my question regarding the system operation in the second reply above or do I have to keep guessing.
Also do you have any bits of circuit showing one of those relays in the activated state. I have never come across relays drawn like that before. I think I can see how they would have to operate but it would be nice to be sure.
Cheers Bob

2 Likes
58

There is no circuit diagram available.

What level were you feeding into it. Could be overload or it could be the quality of audio out of the horn speaker. It is probably not renowned for its HiFi qualities. Also what impedance is it.

  1. I am feeding the “line level” that Sparkfun quote, out of the MP3 Trigger.
  2. Other than that, I will have to measure it with the oscilloscope
  3. The horn speaker is 8 ohm 15 Watt and good quality.
  4. Horn speaker never designed for HiFi.
  5. Horn speaker Characteristics
    SC615 characteristics
    SC615 characteristics1920×2714 392 KB
  6. Horn speaker specs
    SC615 specs
    SC615 specs1127×653 63.2 KB

It might kill the signal level a little bit but to answer that I would have to know the impedance of the audio source. I would personally go for 10K but it should not matter much.

I will change it to 10K a) because I can get a log 10K pot, b) to see what happens.
This one Taiwan Alpha 10k Log 18T Spline 9mm Single Vertical PCB Pot - Altronics

Re Are you going to answer my question regarding the system operation in the second reply above or do I have to keep guessing.
I gave you a thumbs up on that one. Maybe you missed that or maybe it didn’t go through.
Either way you have it correct.

Re relays.
Let me start by saying,

  1. that this information is proprietary and I am not meant to be giving it out. That is why I cannot just give you the whole drawing.
  2. This drawing is very old. and therefore, may be the reason they are drawn that way. I can’t do anything about that.

You asked about relay activated state - see note 2 here

Note 2
Note 21154×260 112 KB

Here are the parts of the circuit that activate the relays
Relay K1

K1 operating circuit
K1 operating circuit1238×500 180 KB

Relay K2

K2 operating circuit
K2 operating circuit1029×195 23.6 KB

Regards
John

1 Like
59

Hi John.
386 amp circuit:

Not surprised. Some suppliers do, some don’t.

Perceived distortion:
The tone itself or how it was recorded and what from are all possibilities. Can’t tell from here.
The oscilloscope will show clipping mainly. Impossible with complex waveform and has to be pretty bad to see anything else but clipping even on a sine wave.

Log pot:
I think you will find volume adjustment far smoother and “nicer” (for want of a better word) by using this.

Relays:
“Contacts are make before break” speaks volumes and explain the drawing of them. They are not all that common and I don’t recall having any dealings, switches yes but not relays. Usual types are break before make. Although of side interest I was not too concerned with the control circuitry but more so what they did and how they did it.

Sstem operation:

There now, not there this morning. Just a yes that is correct is all I wanted to know so I wasn’t chasing fairies all over the place.

This is a bit more serious.
Let ME start by saying. “I have no commercial or employer / employee affiliation with Core electronics apart being a customer”.
You came onto a public Forum seeking help with a problem.
You cannot expect help without providing enough information foe people to attempt to formulate a solution.
If you cannot make this information public and it is indeed proprietary then you are in the wrong place.
This exercise has been looking more and more like a commercial enterprise but I have got interested and carried on. BUT the reference to proprietary information causes me to lose interest rapidly.
Now I don’t know the manufacturer of this system, your employer or what you do there, nor am I interested But I believe this forum is hosted by Core Electronics in support of users of equipment they stock or have an interest in, all customers and hobbyists / makers. If mods and additions to commercial products not stocked by Core are required there are organisations out there that specialise ind providing solutions. I believe Core may have contact details for such a company.

I am sorry but if I have been dealing with confidential or proprietary information I am going to back off.
I think I have provided enough now to allow you to at least try out your idea but if you want to go further I think you are in the wrong place.
Cheers Bob

60

I understand your requirement to work with users of Core products.

This is not a commercial project. I have recently retired and want to get stuck into electronics that I have not had time to do in the past.

Hope we can communicate again in the future.
Regards
John