Seeburg SCH1-4 APU Missing

[ScotK]

OK, I do show connection from the 15 pin connector, pin 10, to the stepper wheel number pins. And the number of connected pins varies based on what number I press. Seems to me like I'm missing a connection from somewhere to pin 10.

[ScotK]

Anyone have a schematic for an APU11? Or an APU10? Pin 10 should be identical on either. I haven't found one yet.


Edit:  Or maybe I'm just misunderstanding the schematic in the SCH1 manual. I guess if it's a wiring schematic it wouldn't have everything on it like gears and stuff.

[ScotK]

Ron, if I'm reading the schematic correctly, there's a switch in the SCH1 APU schematics lableled 2P21 that when in its default normally closed position would connect pins 9 and 10 in the 15 pin connector. That's what would connect the signal cable to the numbers on the stepper wheel.

[Ron Rich]

Scott,
I don't think so-- BTW-- I misspoke earlier--it's 40 VDC applied to the step-up coil, not 25 vac. on the solid state steppers.

[spetersen]

Jan 8, 2020 18:46:42 GMT ScotK said:

Ron, if I'm reading the schematic correctly, there's a switch in the SCH1 APU schematics lableled 2P21 that when in its default normally closed position would connect pins 9 and 10 in the 15 pin connector. That's what would connect the signal cable to the numbers on the stepper wheel.

After looking at the schematic for a while, I think you are correct

[Ron Rich]

Scott K,
I do not have the service manuals here--just going by my great--but sorta foggy, memory-- ! Ron Rich

[scudie]

Hi Scott
Put a piece of wire from terminal 9 to terminal 10.
As you don't have the APU, contact WS is missing, so there is a break in the circuit.
Cheers Scudie

[ScotK]

Thanks for the confirmation everyone. I connected 9 and 10, now every contact on the pulse train is reading 5v with no buttons pressed, as it should, and it should now be working.... except I ran into another road block.

Somehow I burned out the optoisolator. Which makes me think that it's not going to be reliable long term and has me wondering if I should just drop it and connect the signal line directly to the input on the Pi. Or replace the opto with a relay that won't be possible to burn out.

Rather frustrating to say the least. Made me think again if I should just remove the pulse train entirely and use the buttons as raw inputs into the Pi, and convert the entire wallbox to 5v and LED. No points to clean, no pulse train to read, 100% reliable input, lower energy consumption. Also means no going back though.

[scudie]

Glad to hear that you got the selections sorted, There are a few videos on the net / YouTube of raspberry pies connected to wallbox's so it is doable!
Don't know how though.
Cheers Scudie

[spetersen]

If you try to use the buttons directly there will be a huge amount of wiring changes,there is someone on the net that did it that way. I have not had any issues with the optoisolator on the 2 boxes I converted, and they have had heavy use. You may be able to connect it directly, I would check the signal pulse voltage, against the hat, vs it's ratings. (directly to the Pi 3.3v is the limit)and the signal would be inverted from what it is now.(like mine)

You would connect the 5v to the signal wire through a 200 ohm resistor, and connect the signal wire to the input on the hat.

[ScotK]

The hat is tolerant of up to 24v, so no problems there, I can use the 5v directly. I could hook 24vAC up to the pulse train, rectify it to DC, then hook that to the Pi if I'd wanted. That was one reason I picked the hat, it allowed me lots of flexibility when I didn't have a clue yet how I was going to integrate the wallbox. I know it'd be lots of wiring changes but it would be direct input then, so 100% reliability guaranteed.  If things keep giving me problems I may just get irritated enough to switch to that.  

Had another thought... without the APU the numbers and letters are separate on my wallbox. I have one input open yet, so I COULD count numbers and letters on two different inputs. That would make the pulse train easier to calculate, and allow me to troubleshoot which number or letter contact is not working correctly a lot easier. Or even turn letters or numbers off temporarily.

Was thinking about this for the relay:
www.amazon.com/dp/B079FGPC9Y/ref=sspa_dk_detail_5?pd_rd_i=B079FGPC9Y&pd_rd_w=SivdK&pf_rd_p=45a72588-80f7-4414-9851-786f6c16d42b&pd_rd_wg=SDkjT&pf_rd_r=FEKCHZ55495EN337JTRA&pd_rd_r=c172c1d0-11db-4997-b9b9-82c1ca523637&spLa=ZW5jcnlwdGVkUXVhbGlmaWVyPUEyUVRRUUJPOU1STkZCJmVuY3J5cHRlZElkPUEwNTQ3NjI2MVowWEtEWExVT1JWUyZlbmNyeXB0ZWRBZElkPUEwOTc1ODgzMzNVRDU0TkYzTEFGOCZ3aWRnZXROYW1lPXNwX2RldGFpbCZhY3Rpb249Y2xpY2tSZWRpcmVjdCZkb05vdExvZ0NsaWNrPXRydWU&th=1

They have it in single channel version also.

[spetersen]

If you rectify the 24V you will get 35vdc. If you look at the other conversions linked,you will see what people had to do to make it work.
The person that converted to just using the buttons, did so because they made a mistake taking things apart, and couldn't go back.
The issue with using only the buttons would be that you would need 28 additional inputs, or you would need to matrix them in some way.
It could be done with an old USB Keyboard controller, the way they do it on MAME cabinets. The buttons would no longer lock.

Using a relay might work,I would use a small one as it's fidelity would be better, separating the number and letter pulses would complicate things more, the number / letter gap is "huge"(>200ms vs 40-50 ms) compared to the gap between the individual number / letter pulses.

I suggest you read through the conversions in the links to see what they did and why, it's a very good read. I was not able to find a link to the conversion where they used just the buttons, I remember reading it clearly.

The other choice would be the ready made solutions for ~$140

Which side of the opto was burned out?

[ScotK]

How about a solid state relay?

rover.ebay.com/rover/0/0/0?mpre=https%3A%2F%2Fwww.ebay.com%2Fulk%2Fitm%2F372854646796

[spetersen]

That could work, The Omron solid state relay it's based on, has a switching time of 1ms, which seems fine. I am not sure how to connect it, I don't know what the CH1 terminal does.

Ask the seller if it comes with a datasheet or hook up diagram.

This one describes it better

www.ebay.com/itm/1-pcs-One-way-Solid-State-Relay-Module-For-Arduino-P8V8-hgf/124010148661?epid=15019205380&hash=item1cdf94b335:g:clAAAOSwzZFd9Ei2

If the diode side of the opto burned out, the ssr is just as sensitive, if the output side burned out, then this is more robust

The opto that you bought has a second channel, so you can try again

Before you try, put a meter between the signal terminal on the consollette and chassis,With the 5v disconnected, if you see any voltage while selecting, something is wrong.

[ScotK]

I'm going to switch to the second channel of the opto, just worried about long term reliability. I believe the input side burned out. You had mentioned that you had a dropping resistor before your opto input, I was thinking I burned out my opto due to the 5v input even though from what I read I thought it was 5v tolerant. I was hoping the solid state relay was a bit more tolerant, but at least it's guaranteed ok at 5v.