Seeburg SCH1-4 APU Missing

[Ron Rich]

Hi Scott,
You'z welcome !! Ron Rich

[ScotK]

Nov 23, 2019 4:37:59 GMT sooner said:

ScotK - Impressive and most of it way over my head.

It looks like all coins will be treated equally as far as credit goes. Will this system allow for multiple credits to be established? Will you be able to be able to set up different prices for singles vs. albums?

I also noticed that the volume buttons use both series and parallel resistors so total resistance is not proportional to sound level. Will that be a problem?

The coins will all be treated equally.  Since this is for my basement I'm fine with that.  I plan on giving 3 plays per coin drop and having free play available as well.  I'll probably have a tray of coins sitting on top of it for playing songs. I have more free inputs so I could hook up the other two if I wanted to.  

The volume buttons won't be a problem, I hope.  I'm planning on putting 5v to them and just reading the voltage coming out the other end to what would be the speakers.  The voltage should be constant for each button and I'll map that to the volume levels I like on the Pi.  

[ScotK]

Nov 23, 2019 12:19:12 GMT spetersen said:

Does the buffered in 2 supply 5v to the optoisolator? The "signal" at the blue terminal is a pulsed ground with reference to the frame.

The buck converter I used was 35v DC to 5v DC(adjustable for voltage and current) I rectified the 24vac to DC and fed the converter

I don't really show the optoisolator in the drawing completely because I don't have it fully worked out yet, mostly because it's still on its way here.  I was thinking to take a 5v feed from the power supply to the optoisolator which would, when energized by the pulse train, allow the 5v through to the input on the Pi.  

The 5v supply I ended up going with is this: 

www.amazon.com/gp/product/B00RE6QN4U/ref=ppx_yo_dt_b_asin_title_o01_s00?ie=UTF8&psc=1

It's 3amp so I may have to add a second one in parallel.  They're not much bigger than my thumb in length and width so I can easily add a second if needed.  

[ScotK]

Partially assembled... I mounted a piece of plexiglass via double sided foam tape to the back wall of the Seeburg to mount the Pi to, and hooked into all the connections I need with the 15 pin connector.  Much thanks for the assist there Ron, that looks nice.  

Attachments:

[sooner]

Nov 25, 2019 15:40:23 GMT ScotK said:

Nov 23, 2019 4:37:59 GMT sooner said:

ScotK - Impressive and most of it way over my head.

It looks like all coins will be treated equally as far as credit goes. Will this system allow for multiple credits to be established? Will you be able to be able to set up different prices for singles vs. albums?

I also noticed that the volume buttons use both series and parallel resistors so total resistance is not proportional to sound level. Will that be a problem?

The volume buttons won't be a problem, I hope.  I'm planning on putting 5v to them and just reading the voltage coming out the other end to what would be the speakers.  The voltage should be constant for each button and I'll map that to the volume levels I like on the Pi.  

Yes the voltage will be constant for each button but the voltage may not be proportional to the volume. For example, when I measured resistance across the button circuit it was zero on hi. Resistance was higher for both the low and medium settings but... the middle button had a higher resistance than the low. I assume the voltage you measure will be similarly disproportional. If that's a problem it would be easy enough to change some resistors in the button circuit. You can also set up a 4th volume setting for when no buttons are pressed. That can be useful to give more volume range or to mute the box when using Bluetooth so you don't get the echo delay effect.

[spetersen]

I think it would be easier to access the ports and GPIO if the Pi were mounted horizontally behind the right speaker. Do you plan to connect a monitor for debugging?

[ScotK]

Nov 25, 2019 19:27:41 GMT sooner said:

Nov 25, 2019 15:40:23 GMT ScotK said:

The volume buttons won't be a problem, I hope.  I'm planning on putting 5v to them and just reading the voltage coming out the other end to what would be the speakers.  The voltage should be constant for each button and I'll map that to the volume levels I like on the Pi.  

Yes the voltage will be constant for each button but the voltage may not be proportional to the volume. For example, when I measured resistance across the button circuit it was zero on hi. Resistance was higher for both the low and medium settings but... the middle button had a higher resistance than the low. I assume the voltage you measure will be similarly disproportional. If that's a problem it would be easy enough to change some resistors in the button circuit. You can also set up a 4th volume setting for when no buttons are pressed. That can be useful to give more volume range or to mute the box when using Bluetooth so you don't get the echo delay effect.

As I read the schematic, the way I have it connected if no buttons are pressed then the voltage should be zero.  So I'll use that reading to mute the jukebox.  The only thing I care about is that I get a constant and similar reading each time the same button is pressed.  When I set it up I'll read the voltage with the low button pressed and set that to the low volume on the Pi, then the same with the medium and high buttons.  The only thing I'm using the buttons for at all is to give me a voltage reading so I know which volume button is pressed.  Actually setting the volume will be done in software.  

That volume function is one reason I needed the automation hat for the Pi.  A regular Pi has inputs and outputs.  The hat has the relays and an analog connection that will just give a reading on the voltage.  

[ScotK]

Nov 25, 2019 23:47:27 GMT spetersen said:

I think it would be easier to access the ports and GPIO if the Pi were mounted horizontally behind the right speaker. Do you plan to connect a monitor for debugging?

I thought so too but i needed that space for the 24vAC transformer.  I wasn't planning on connecting a monitor.  If I do I'll have to use a ribbon cable as there's little room beneath the Pi between the HDMI and the wallbox chassis.  I'll have remote access to the Pi via SSH as it's running Linux.  The only cause I should ever have to connect a monitor is if it's not available remotely for some reason.  That's one reason I'm using one or more of the light bulbs as status indicators too.  I could hook a relay to the song catalog light and turn that on via the relay when the Pi is booted and ready to run.  That way I know it's ready.  Also tells me that something is wrong if I power it up and the light never comes on.  

[spetersen]

My red / blue led is the same thing, if it does not light it isn't booting, Where is your music going?

The Hat may be a little close to the cams in the title strip mech

[ScotK]

Nov 26, 2019 4:30:29 GMT spetersen said:

My red / blue led is the same thing, if it does not light it isn't booting, Where is your music going?

The Hat may be a little close to the cams in the title strip mech

I had the Pi and hat in there already when figuring out how to arrange it to fit all the components, it'll fit.  It's a tight fit though.  

I've decided to flip the Pi upside down after thinking it over.  That'll give me unfettered access to the HDMI and USB as well as better access to the hat terminal blocks. I just have to watch the cams when wiring to make sure they stay away from it.  

[spetersen]

There is a possibility you may have to place a small cap across the relay outputs to reduce switching spikes, as this can cause some pi inputs to false trigger, I had a problem with this on my skip song program ,it would trigger when adding credits sometimes.
Nice project!, I'll be watching

[ScotK]

Nov 26, 2019 19:18:05 GMT spetersen said:

There is a possibility you may have to place a small cap across the relay outputs to reduce switching spikes, as this can cause some pi inputs to false trigger, I had a problem with this on my skip song program ,it would trigger when adding credits sometimes.
Nice project!, I'll be watching

I don't think I'll have to worry about switching spikes, the relays aren't controlling power to anything that will be going to inputs.  All the inputs are straight from the +5vDC supply.  The relays are only connecting pins to ground in both cases on 24v lines to simulate the APU or turn on a light bulb.  

Thanks!  You, Scudie, Sooner, and Ron have helped a lot so far.  Ideas, feedback, even part numbers.  I wish I had this kind of collaboration at work.  

[spetersen]

I guess the motor would be the only one that could be under load.

[ScotK]

Nov 23, 2019 4:37:59 GMT sooner said:

ScotK - Impressive and most of it way over my head.

It looks like all coins will be treated equally as far as credit goes. Will this system allow for multiple credits to be established? Will you be able to be able to set up different prices for singles vs. albums?

I also noticed that the volume buttons use both series and parallel resistors so total resistance is not proportional to sound level. Will that be a problem?

Forgot to answer your other question about the pricing for singles vs albums.  I'm hooking all three coin drops to the same power source, but if I put resistors in between so I get differing voltages from each coin drop, and if I moved the incoming wire to an ADC input on the Pi so I could sense the differing voltages from each type of coin, then I could know exactly how much money has been dropped in.  Since the Pi is taking the input and simulating the APU I could in theory set it up in the software on the Pi to require different credits for any number of different selections, and each selection could be an album or a single.  Once at least one coin has been dropped the Pi would ground the single credit wire allowing the pulse train to run to input the selection to the Pi.  The Pi would only play the single or album once enough credits are sensed by reading the coin drop switches.  I could also make use of the album selection light to indicate when not enough credits have been deposited.  

Indeed, when I designed all this in my head I wondered if there would be any money in doing several of these and selling them after being retrofitted.  My cost would be under $200 for parts + whatever cost is on the Seeburg consolette now that I have everything worked out.  That's assuming the consolette is functional too.  

[ScotK]

Nov 26, 2019 20:45:33 GMT spetersen said:

I guess the motor would be the only one that could be under load.

I think so, yes.  And I've decided I'm not going to control the 24vAC feed with the Pi.