Wilba

Directly plugging the unbalanced stereo output of sammichSID into a mono balanced input is not going to work. You need an adapter. I suspect you may be directly wiring the TRS plug (into the sammichSID) to an XLR plug (or a TRS plug) and plugging it into an XLR mono balanced input (or a TRS mono balanced input). That is definitely NOT going to work. Balanced inputs have three contacts/pins/wires but it's NOT stereo. A single balanced mono input is a two-wire pair plus ground, and the two wires are the same mono signal just inverse of each other. I have one balanced input in my sound card (M-Audio Delta 1010-LT) so I made up an XLR adapter at the other end of the cable which ties one balanced input (the negative one I think) to ground, the other to the signal. This works because the input measures the relative difference between the two inverse signals, so if one is ground and the other is the unbalanced signal then it still works. Also, you CANNOT plug a mono plug (TS) into the sammichSID's stereo (TRS) socket. You can't get stereo out of a mono plug or a mono cable. I assume your SIDs are not damaged, and the problem is just the plugs/wiring to whatever you're using. The simple way to test this is to just get a cable with stereo TRS plug and two male RCA plugs on the other: and plug it into a stereo amp or your TV or something. They often are sold with the 3.5mm plug, so get an adapter for that end. Alternatively, make your own stereo TRS plug to two mono TR sockets adapter. Or a stereo TRS plug to two mono TR plugs. In case you need more info: http://en.wikipedia.org/wiki/TRS_connector

Do you have two SIDs installed? Are you using a mono plug in the sammichSID's stereo socket?

I can't explain why that PSU doesn't work. Some switchmode PSUs I've tried work OK but are noisy. I've only found one that works without noise, looks like the one you have, but it takes a 3-pin power plug (like on a PC) and since tip is positive, I use the shunts in JBP to bypass the bridge rectifier so negative of the PSU is also ground of the sammichSID. This might explain why there's no noise - it's even less noisy than a regulated linear (transformer) plugpack.

I'm a bit confused about how that can happen. The red LCDs are "high-powered" so you should use a shunt in JR4 and JBL set to "12V" (i.e. input power) and set brightness pot to half-way. Can you tell me what happens if you set the brightness pot to minimum? I suspect the problem is related to current, perhaps your PSU cannot deliver enough current when the brightness is maximum. Also tell me what happens if you remove the SIDs. Describe what kind of SIDs and PSU you are using, the voltage, current rating, regulated or unregulated, etc. I would like to "blame" the PSU for this behaviour, but it could be just that the sammichSID is drawing too much current for the PSU.

Batch #3 is being prepared now. Due to cancellations, people deferring to later batches etc. I have only 25 confirmed orders, and still waiting on people to reply to my emails, so chances are good that if you email me to go on the waiting list, you will get a kit from this batch in September. Based on the low popularity of sammichFM (which I expected, it's not as retro-cool as a SID synth which is comparable to expensive commercial products like SidStation and HardSID), Batch #4 might not happen until December or even January, it might take that long to get 50 orders, and I'd probably want at least 60 because there's inevitably a 30% or more cancellation/postpone rate. SO many people go on my waiting lists and then bail out when it's time to order or pay, which is difficult to manage sometimes, but probably for the best, because then kits go to the people who really want one and know what they're doing ;-) TL;DR? PRE-ORDER NAO!

Don't throw it in the river!!! Send it to nILS for proper disposal. Or keep at it and fix it. BTW I noticed I have not uploaded the control surface PCB layout as PDF. It's up now, and referenced in the wiki: http://www.mb6582.org/sammichFM/sammichFM_CS_PCB_Rev_1.pdf

All the LCDs I sell now are high-power LCDs and require a shunt in JR4... otherwise the LCD only gets low current (max. 25mA) which is only good for edge-lit LCDs like some white-on-blue LCDs or the amber/true green ones I was selling in the first few batches (before I switched to standard yellow-green LCDs).

If it's formatting ALL the banksticks then there must be an issue writing to the bankstick... i.e. it's reading back zeroes all the time so the firmware wants to format them, which would also explains why patches aren't storing. I suggest taking out all the 24LC512 and only inserting one in "Bank 1" socket. Then check the voltages around there (as per the voltage check section of the build guide). Make sure the pins that should be ground are really ground (those pads can be tricky to solder). Then check there are no shorts or bad connections for the data pins going between the PIC and the banksticks. There are two data lines which they all share, and which are also connected to J4 (MD and SD pins). See the tracks using the PDF: http://www.mb6582.org/sammichSID/sammichSID_Base_PCB.pdf

Hole numbers matter. 1st and 2nd pins are connected. 5th and 6th pins are connected. So you can use capacitors with 0.1" or 0.2" lead spacing. It makes more sense if you look at the PCB: http://www.mb6582.org/sammichSID/sammichSID_Base_PCB.pdf I should have just put photos in the build guide with photos of the two possible ways of plugging in caps.

The 1st patch is probably still the "Init" (default) patch, which shouldn't be quiet. Possibly you have Osc 1 quieter than the others. Try initializing the 1st patch again (via the Shift menu) Then compare the volume when playing each Osc with the same settings. i.e. goto Osc menu, set Osc 1 to off, set Osc 2 to Pulse. Repeat with Osc 2 off, Osc 3 Pulse. This is just a theory. Perhaps there is some hardware problem too. Check the filter caps are correct for the SID type and installed correctly. Also, if you have two SIDs, is this problem only happening on one SID or both?

OK I'll assume you installed SIDs and the 24LC512 (banksticks) before testing the control surface... not that that's a bad thing but I recommend in the build guide to get everything working and test the buttons/knobs etc. before you go the next step and install the SIDs and banksticks. Go here: http://www.midibox.org/dokuwiki/sammichSID and scroll down to sammichSID Troubleshooting and download the test app. Upload it to sammichSID just like the MB-SID V2 app. Report back with results of that test. Assuming LEDs are working, then most likely it's not a power issue, you say you did voltage checks already so the ICs should be working. So it might be a bad connection on the SI pin between the PCBs, or SI is shorted with something, or ICs in the wrong place or installed backwards, or the first 74HC165 in the chain is dead (try swapping them and do the tests again).

Triangle waveforms don't sound as loud as others. Also patches using the filter might not be as ones not using the filter. Maybe your filter is sounding mushy or not working well? Identify which patches sound less loud, try turning off the filter, etc. Or if you want, upload some wavs of the same note played in some patches which you call loud and which you call quiet (and don't touch the recording level in between recording those patches)... then we can compare with other MIDIbox SIDs.

Actually you should swap them with me for 8x 6582A SIDs... it's not called MB-8580R5 now is it? :D

Were the buttons and knobs working at some point, and then stopped working entirely? Did you already install the banksticks and uploaded the default patches? Just thinking perhaps it is set to a sequence patch and note events are triggering the sequence. From memory, the last default patch is Neo Zelda or likesuchas, and immediately after uploading the 128 patches, the "current" patch is the last patch. Or it could be set to some other patch. What is it showing on the LCD?

6581 SIDs get hot... not much you can do about that, apart from stick on some heatsinks if you want (see other threads for ideas). With a regulated 12V DC input (required for 6581!) then the heatsinks and ventilation should be enough. Perhaps warming up caused a nearly bad solder joint to go fully bad, or something. One other thing to check... if you're using a "high-power LCD backlight", be sure use JBL set to 12V, shunt in JR4, and turn the brightness pot to half-way or less. You would not want JBL set to 5V, as this will only add more heat to the 5V regulator, which is under enough strain as it is, supplying those power-hungry 6581 SIDs.

@brogram: what are you using to test MIDI? Do you get the same results when just using a MIDI cable to connect PC's MIDI In to MIDI Out? What is strange here is you say you got messages from the PIC, i.e. the "upload request" messages, so the sammichSID's MIDI Out is working. Try the short between JMI and JMO middle pins, but with sammichSID MIDI In disconnected. i.e. test if you are getting garbage data from the MIDI In circuit when there is no MIDI going into the sammichSID. It's possible you have a bad solder joint (or short) somewhere in the MIDI In circuit, i.e. around the 6N138, the resistors, etc. perhaps the 6N138 has bad ground joint or bad +5V joint.... perform the voltage check tests from the build guide again to check it is all good.

If you connect middle pins of JMI and JMO then yes, you should expect MIDI loopback... events going into the sammichSID via MIDI In should be echoed back out the MIDI Out. Test your MIDI works by actually looping back the cable, i.e. connect MIDI In to MIDI Out, run some tests. Run same tests using the sammichSID. If you don't get MIDI echoed back then the problem is perhaps the 6N138.

Designators (like R5/R6) are the same. Headers are slightly different, and it's pretty clear which is the PIC and which is the 6N138. That diagram linked to in the previous post shows the corner of the PIC and which pins they are. or if it is particularly stubborn, It should, but it would be tricky to maintain while driving MIOS Studio. If you have thin bendy wire, cut two lengths, wrap each around the pin first, then twist together. Or use alligator clips to clamp on the middle pin (without shorting with the adjacent pins).

Did you do all the MIDI troubleshooting here? http://www.ucapps.de/howto_debug_midi.html Especially do: which is easily done by removing the PIC and inserting some thin wire (like a cut resistor lead bent into a U-shape) into the Tx and Rx pins. Alternatively, leave the PIC installed, remove shunts from JMI and JMO and connect the MIDDLE pin of JMI to the MIDDLE pin of JMO. This disconnects the PIC from the MIDI In/Out circuits and loops-back the MIDI In signal to MIDI Out. You need to get MIDI loopback working in order to prove the 6N138 is working, as well as your MIDI interface, MIOS Studio setup, drivers, etc. i.e. prove that the PIC is receiving MIDI but not actually working. If the 6N138 is not working then it is easily replaced (by me, if you can't find one quicker yourself).

Regarding MIDI In: - check 6N138 is installed - check shunts in JMI (and also JMO) It's probably one of these, if not, I'll help more later (well, you can start by checking solder joints at JMI/JMO, the socket for 6N138, the DIN sockets etc.) Regarding voltage: I'll assume you haven't installed the 6581 yet. 12.44V is not too high. Regulated power supplies aren't perfect. That voltage might drop after installing SIDs, but even if not, it's fine, you won't fry the SIDs. Since the LCD is working and it boots, I can assume 5V supply is good, so the 12.44V would have nothing to do with MIDI In not working.

I will cease and desist forthwith.

Great track unikoi! :thumbsup: :thumbsup: :thumbsup: I like those 5ths (if that's the right way to say it, I'm a music newb). Reminds me of this (at 1:31): P.S. seems to be a new MP3 plugin on the forum... no volume control... can't download the MP3 (unless you are admin and can view/edit the post text). Maybe you should upload it to Soundcloud too.