Trying to read a schematic

elmegil

I bought a non-functional Roland S-10 for a very cheap price, with the intent of learning stuff while working on it. Originally I had considered stripping it, but then decided I would try to get it to work and then make up my mind what to do with it.

I have the service notes (thank you Roland corporation!), and have looked them over.

I have a high pitched whine when (what looks to be) the backlight for the LCD is plugged into the power board. Unplug, whine goes away. No light ever.

It's a 3 position connector with two pins connected. When I measure the voltage on them I get 24V DC and 136V AC. That sure doesn't seem right. Smile

I thought I had measured C19 as being wrong, so I replaced it, but I'm guessing I just wasn't able to measure it accurately in circuit because it's measuring just fine on its own.

What's confusing me though is the schematic diagram (see attachment for the section). I don't get what the two lines with the marking "1 2 3 4" are supposed to be. There's no chip there. it doesn't look like a normal bus notation. So I'm at a loss to understand what that is so that I can then measure at the circuit and see what it's supposed to do.

BTW that's 5.25V (digital supply) into the coil L2 and digital ground on the negative side of C18 and also to pin 1 of the connector.

Anyone recognize this?

Thanks

Screen shot 2012-07-13 at Jul 13 10.37.53AM.png

Description:

circuit from power to LCD

Filesize:

58.35 KB

Viewed:

1146 Time(s)

Screen shot 2012-07-13 at Jul 13 10.37.53AM.png

elmegil · Posted: Fri Jul 13, 2012 9:35 am Post subject:

Ah, there IS a transformer there (very small one), which would line up with T1. Not sure why that wasn't coming through my brain, let me pursue that....

Ok, so it's an inverter circuit, DC back to AC, I'm still not sure how to read the diagram, but I can ohm it out.

elmegil · Posted: Fri Jul 13, 2012 1:53 pm Post subject:

And here we go, from the spec sheet for an inverting transformer, essentailly the same circuit except for an LC filter on the front end.

Screen shot 2012-07-13 at Jul 13 3.47.06PM.png

Description:

Filesize:

17.8 KB

Viewed:

1089 Time(s)

Screen shot 2012-07-13 at Jul 13 3.47.06PM.png

Blue Hell · Posted: Fri Jul 13, 2012 2:26 pm Post subject:

bobine .. ignition coil in english probably? never had seen it drawn like that before either. Anyway, looks like one single coil with a couple of taps.

elmegil · Posted: Fri Jul 13, 2012 2:49 pm Post subject:

Here are the things I found explaining it

http://www.butlerwinding.com/store.asp?pid=28348

http://www.elytone.com.tw/productDetail.asp-ParentID=101&PageID=EEL-10001-1.html

elmegil · Posted: Sat Jul 14, 2012 10:59 am Post subject:

Next S-10 challenge, checking things on the output board... there's a "big" LC filter (as in bigger than an IC, not huge) with three stages. I'm not having any luck finding information about how these are used, although I have a very clear circuit diagram on the schematic itself.

Anyone have any resources for these?

elmegil · Posted: Sat Jul 14, 2012 6:13 pm Post subject:

And what kind of crystal circuit only uses one capacitor?

This is driving an FPGA that handles the ADC/DAC logic for the S-10.

I'm not seeing a recognizable signal there (yet? Need to try a couple more things), so I'm partly wondering if the lack of ability to digitize is due to a problem here. When I saw that there was only one cap I first thought this was the issue, but from the schematic it's clear that this was deliberate.

Screen shot 2012-07-14 at Jul 14 8.06.47PM.png

Description:

Filesize:

45.28 KB

Viewed:

948 Time(s)

Screen shot 2012-07-14 at Jul 14 8.06.47PM.png

Blue Hell · Posted: Sat Jul 14, 2012 6:23 pm Post subject:

The capacitor config used depends on the actual xtal driver circuit in the chip - the chip datasheet should have something on that.

Still you should be able to see a clock on either x tal 1 or x tal 2 (the one not showing a clock may crash the thing when you hit it with a probe, that would be the feedback input, but the driver side should be somewhat solid, solid enough to drop a probe in) .

elmegil · Posted: Sat Jul 14, 2012 7:10 pm Post subject:

Thanks for that, Blue Hell. Switching legs of the crystals I've been able to observe signals.

I think I have a scope limitation (I'm using one of the small cellphone sized DSO Quad scopes designed by Seeed studios). There are three crystals on the main board; one 6.5MHz is for the QD Driver and USART used for MIDI, one is 12MHz for the main CPU (a resonator rather than a crystal proper), and one is this one, at 26.880MHz. My scope is reading the first two as 6.5KHz and 12KHz. The last one is supposed to be 26.880MHz, but I'm reading it as just under 9KHz. That would be a factor of 3 (I'm rounding a bit, the actual frequency was 8.something) as well as a factor of 1000, so that seems fairly likely to be an aliasing issue with the scope.

I just realized too that this same FPGA controls access to the DRAM, and while I can't get past the quickdrive diags until I have a disk to put in (on its way), the DRAM diags are before that and they pass. So the FPGA is at least somewhat functional, so there's probably not a clock problem.

elmegil · Posted: Mon Jul 16, 2012 10:10 pm Post subject:

Turns out most of the beast seems to be working fine. The driver for the EL backlight still whines and no backlight comes on, so I suspect a problem with the EL itself. Not a critical part, I'm going to leave it disconnected to avoid the whine. I have had NO luck trying to modify anything behind an LCD bracket in the past.

I've been able to sample and play back.

I've been able to use it as a MIDI controller for other keyboards.

I've been able to demonstrate that MIDI IN goes to MIDI THRU. Hm. Ought to go back and verify that MIDI IN will play the keyboard when I have a sample loaded.

I have NOT been able to get SysEx messages to dump samples to the S-10, but I suspect this is user error/badly documented process to make that work. Just posted a question about it over in the Sampler section.

I have QD disks in my hot little hands, but the drive belt is, predictably, melted, so I have a couple of those on their way (inexpensive, always good to have a backup). Once I get that done I hope to get past the QD diagnostics and get the real MIDI diags done just to be 100% sure.

Need SysEx, need QD, need to blow out dust and clean the keys. All told I'm pretty happy given what I paid for it and what I've seen working units going for.