Open Source release of MiSTer Multisystem Console hardware all-in-one I/O design

[DeKay]

I wanted to share my build progress, this is where I’m currently at after ordering all parts from the BOM @emiliom posted. I only had an issue with 2 or maybe 3 items that were out of stock but alternates were available. I’m located in the US and I am still waiting for 2 parts: USB IC and the Vertical HDMI connectors. I am having trouble ordering the optical toslink audio connector from the provided site for whatever reason, if I decide I need it I’ll just give them a call and order manually. Sadly without the USB IC I don’t think I can test my setup any further. So far I’m not seeing any LEDs blinking in the back next to the Network port (I assume that’s what they’re for?) it does have a Network symbol on the Mister menu that pops up and shows the date/time so maybe that’s a good indicator? One thing to note was the 2 40 some odd pin connectors that interface the DE10 nano stick out a Ton on the bottom of the board after being placed and will need to be individually snipped which was slightly annoying (wear protective eyewear).PXL_20230821_185251139.jpgPXL_20230821_185258054.jpg

Exciting!

LED9 and LED10 on the MultiSystem board are driven by the USB Chip signals LED_5 & LED_6 and show the status of USB Ports 5 & 6 on that chip. See clip from the schematic below The LAN port LEDs on the DE-10 are not carried over that little extension cable to the back of the MultiSystem. You would have to look on the DE-10 itself to determine network activity.

I would think that you could disconnect the USB patch cable between the DE-10 and the MultiSystem and just plug one USB peripheral directly into the DE-10 if you want to test further before waiting for that chip to arrive.

That the header pins were way too long and stuck out below the board just suggests that the wrong type of header pins were specified. Normally these will be just long enough on one side to fit through the PCB.

Screenshot_20230821_141634.jpg

[rezendes]

Just got the vertical HDMI connectors in and I can now verify the height of the pin connectors for the de10 nano interface (ones with super long pins coming out the bottom) is correct and allows the hdmi dongle to fit nicely. Now I’m only really waiting on the USB IC, I won’t need the SCART or TOSLINK. I am curious though if there’s any way I can verify my installation all works correctly such as Ram / crystal oscillator…

[multisystem]

Great job on the Multisystem build. Really nice to see you have a running system and an almost fully populated PCB.

The LED's 9 and 10 are the back two USB ports, so they will only light when the USB Hub is fitted and also when a USB device is connected to the back USB ports.

The Ethernet is just a straight pass-through, so if you have the network symbol under the menu and the time is being displayed, you have network access and it's working.

There are no external Network LED's on the Multisystem, they are on the DE10-nano board.

Good luck with the rest.

I wanted to share my build progress, this is where I’m currently at after ordering all parts from the BOM @emiliom posted. I only had an issue with 2 or maybe 3 items that were out of stock but alternates were available. I’m located in the US and I am still waiting for 2 parts: USB IC and the Vertical HDMI connectors. I am having trouble ordering the optical toslink audio connector from the provided site for whatever reason, if I decide I need it I’ll just give them a call and order manually. Sadly without the USB IC I don’t think I can test my setup any further. So far I’m not seeing any LEDs blinking in the back next to the Network port (I assume that’s what they’re for?) it does have a Network symbol on the Mister menu that pops up and shows the date/time so maybe that’s a good indicator? One thing to note was the 2 40 some odd pin connectors that interface the DE10 nano stick out a Ton on the bottom of the board after being placed and will need to be individually snipped which was slightly annoying (wear protective eyewear).

EDIT

I wired up a USB OTG cable and keyboard is working and update all script is downloading files but still no network LEDS (#9 and #10) assuming that’s what they are for…. Is it possible that the USB IC is somehow necessary for the power/operation of those LEDs like maybe they wired it funny?

[emiliom]

So I finally got all the surface mount components soldered. After cleaning all the flux off the board I noticed I had a bridge across 2 pins of the left hand memory chip, and what looks like a slightly bent pin on one of the HDMI connectors. So I'll fix those tomorrow and then get on with soldering the through hole components.

SMD Complete.JPG

[emiliom]

Phew, long day. And my throat seems to have the permanent taste of Flux remover in it. Here's my fully populated completed board. I think I better wait a couple of days to test it. I know these chemical cleaners dry fast, but I'm not sure about what happens when they get inside the header / connector sockets so best be safe. I'm fairly confident in most of the soldering, but I'm not sure about the horizontal HDMI connectors. They gave me a lot of grief. I guess I'll find out in a couple of days.
To make the SCART socket fit I cut the plastic plugs on the rear of the connector down with a dremel type tool until they were flush with the PCB, then smoothed the edges so they wouldn't be sharp. That way if someone jiggles the SCART plug about it won't damage the board, and there should be less strain on the solder joints as each plug will stop the socket from 'lifting up' at the opposite end. At least that's the theory.

Board Complete_Front.jpg

Board Complete_Rear.jpg

[emiliom]

One thing to note was the 2 40 some odd pin connectors that interface the DE10 nano stick out a Ton on the bottom of the board after being placed and will need to be individually snipped which was slightly annoying (wear protective eyewear).

I found this myself today. The ends definitely fly off a some speed when snipped! I saw Lee found a Samtek alternative, ESQ-120-23-T-D, but it's a bit pricey compared to the SparkFun, at least on Mouser. When I get some time I'll see if I can find a SparkFun version with shorter pins that's cheaper.

[LeeW]

Your board is looking fancy and clean emiliom.

I finished my board as well, but unfortunately none of the USB ports work, any ideas/suggestions on how to track down the issue? The DE-10 board connects to the PC via USB when powered, so the issue is on this board. Could this be due to damage to the 40 pin header when desoldering it?

5sWj3Oe.jpg

lyBrdNK.jpg

As I mentioned before, the original 40 pin headers I got were too short. I left the SDRAM header and just cut that module from the PCB instead. But I didn't have any choice with the bottom one and had to desolder it. This was a nightmare. I spent ages trying to remove the solder using desoldering braid and it was just not working at all. In the end I ended up destroying the shroud

CaWpGyw.jpg

which then made it fairly easy to remove the pins individually. Ended up buying the Engineer SS-02 solder sucker to remove the last bits of solder and was impressed how good a job it did. Lesson learned, sometimes you just need the right equipment (and also patience, which I admit I ran out of when handling this haha).

[Missus]

Your board is looking fancy and clean emiliom.

I finished my board as well, but unfortunately none of the USB ports work, any ideas/suggestions on how to track down the issue? The DE-10 board connects to the PC via USB when powered, so the issue is on this board. Could this be due to damage to the 40 pin header when desoldering it?

As I mentioned before, the original 40 pin headers I got were too short. I left the SDRAM header and just cut that module from the PCB instead. But I didn't have any choice with the bottom one and had to desolder it. This was a nightmare. I spent ages trying to remove the solder using desoldering braid and it was just not working at all. In the end I ended up destroying the shroud (yikes) which then made it fairly easy to remove the pins individually. Ended up buying the Engineer SS-02 solder sucker to remove the last bits of solder and was impressed how good a job it did. Lesson learned, sometimes you just need the right equipment (and also patience, which I admit I ran out of when handling this haha).

Could it be the usb cable connecting the de-10 to the MMS?

[emiliom]

Your board is looking fancy and clean emiliom.

Yeah, let's see if it works now! I have nightmare visions of having to redo my HDMI connectors!

I finished my board as well, but unfortunately none of the USB ports work, any ideas/suggestions on how to track down the issue? The DE-10 board connects to the PC via USB when powered, so the issue is on this board. Could this be due to damage to the 40 pin header when desoldering it?

I don't think desoldering the 40 pin connector would affect it as I don't see any USB connections on either of the 2 40 pin connectors.
The first 3 things off the top of my head are; bad USB cable (as Missus suggested), USB port JP11 bad solder joint / faulty, or problem with FE2.1 chip?
Cable is easy enough to swap out to check. If you look at the circuit diagram for the FE2.1 connections in Heber's docs (https://github.com/Heber-co-uk/Multisys ... /23005.pdf) I guess you could do a continuity check between the pins designated for each USB port to each USB port. Then you could eliminate bad joints? Might not be easy to do though given that the USB port connections are on the underside of the board. But if the connections are ok and the cable is ok, maybe swap the FE2.1 out?

As I mentioned before, the original 40 pin headers I got were too short. I left the SDRAM header and just cut that module from the PCB instead. But I didn't have any choice with the bottom one and had to desolder it. This was a nightmare. I spent ages trying to remove the solder using desoldering braid and it was just not working at all. In the end I ended up destroying the shroud (yikes) which then made it fairly easy to remove the pins individually. Ended up buying the Engineer SS-02 solder sucker to remove the last bits of solder and was impressed how good a job it did.

Wow! You really went full Jack the Ripper on that 40 pin connector! Interesting what you say about the braid because I was thinking the same thing initially. I have this toolkit I've had for years and years which is a sort of general purpose thing, but has pretty much everything in it, including a roll of wick that I didn't even realise was in there. So I'd already bought a small roll myself of MG Chemicals Superwick from Amazon (https://www.amazon.co.uk/gp/product/B008O9W2FA/) just in case. I'd never used wick before so I had no idea about it other than watching others on YouTube use it. Anyway, the first time I had to take a bit of solder off I used the wick that came with my old toolkit...and it was rubbish! It took off sod all! So then I also started using a de-soldering pump that came with my soldering station, and that worked pretty well. Bit tricky sometimes depending on the situation, but generally good. But then I got into the horror scenario of accidently putting solder on 2 pads for a single resistor, and the desolder pump just doesn't get it all off. I was never going to get that resistor on without wrecking it if I couldn't put in down flat on the board, so out of desperation I tried the wick I'd bought on Amazon to remove the solder...and it got rid of it completely, and really easily! I think it really, really, comes down to the individual wick. Some of it's just crap but some seems to work well.

Lesson learned, sometimes you just need the right equipment (and also patience, which I admit I ran out of when handling this haha).

Good advice for us all!

[DeKay]

I finished my board as well, but unfortunately none of the USB ports work, any ideas/suggestions on how to track down the issue? The DE-10 board connects to the PC via USB when powered, so the issue is on this board. Could this be due to damage to the 40 pin header when desoldering it?

As emiliom suggested, a continuity check is a good start. You could unplug the power cable, pull off the Nano, and check continuity between the USB connector pins on the Nano to the USB connector pins on your board. Check continuity on the data lines all the way to FE2.1.

With the unit powered on and the Nano back on, do any of the LEDs light when plugging in to the port or do they always stay off? Are you able to check that the FE2.1 Is getting the right voltage at each of its power pins (VDD_* on the schematic around the chip).

Another thing I would check is U4. If its FLAG- pin is at or around 0V, it is sensing something wrong. Also check that USB_EN is around 5V. If it isn't, then it isn't putting out power to the USB ports and that will ruin your whole day.

It goes without saying but closely inspect the area around U4 and U5 for shorts and opens.

One last thing - I know there was some earlier discussion on what was the right crystal to use for the 12 MHz X2. If you had a scope, you could check if you saw oscillation on FE2.1 XOUT or not.

Good luck.

[LeeW]

The cable works when connecting to pc, and I've tried a few cables. Don't get any USB LEDs at all. Although there appears to be continuity. All of the 5V lines and 1.8V lines appear to be correct (USB_EN is 5V).

Problem seems to be that all of the 3.3V lines are showing as 4V. XIN is 3.3V but XOUT is 4V. So... the crystal is bad?

Anyway, the first time I had to take a bit of solder off I used the wick that came with my old toolkit...and it was rubbish! It took off sod all! So then I also started using a de-soldering pump that came with my soldering station, and that worked pretty well. Bit tricky sometimes depending on the situation, but generally good. But then I got into the horror scenario of accidently putting solder on 2 pads for a single resistor, and the desolder pump just doesn't get it all off. I was never going to get that resistor on without wrecking it if I couldn't put in down flat on the board, so out of desperation I tried the wick I'd bought on Amazon to remove the solder...and it got rid of it completely, and really easily! I think it really, really, comes down to the individual wick. Some of it's just crap but some seems to work well.

My understanding/experience is that it works well for cleaning up SMD pads but not very well for removing THT components. Also you can apply flux to the braid and a little bit of solder to improve it's performance.

[emiliom]

Does anyone know how long the internal Ethernet and USB cables that connect to the DE-10 are supposed to be? The multisystem manual just says the included cables are “short”.

[emiliom]

Problem seems to be that all of the 3.3V lines are showing as 4V. XIN is 3.3V but XOUT is 4V. So... the crystal is bad?

If the Crystal was bad but a signal is getting from XOUT to XIN, wouldn’t that just mean the timing of the controller would be off? Or does a wrong voltage at XIN stop it working completely?

My understanding/experience is that it works well for cleaning up SMD pads but not very well for removing THT components. Also you can apply flux to the braid and a little bit of solder to improve it's performance.

Mine came with flux impregnated so I guess that helps. It really seems to absorb solder very well. I found it useful for the pads but also if I made a blob bridge between 2 pins on the big connectors. It did a good job of absorbing the excess.

[DeKay]

Problem seems to be that all of the 3.3V lines are showing as 4V. XIN is 3.3V but XOUT is 4V. So... the crystal is bad?

Wait. Whoa. What? "All" the 3.3V lines are showing as 4V? That is really bad if you mean that 3.3V is measuring 4V across the entire Multisystem board. That is far enough out of whack to do some damage. I think that that voltage is sourced from the DE-10. I'd start by pulling it off to see if the DE-10 3.3V actually reads 3.3V when not connected to the Multisystem. Is the 3V3_MEM voltage (see schematic Page 1 JP1 Pin 29) also reading 4V?

Don't go any further until you sort this out. Then worry about the USB. In fact, perhaps it is the 4V supply that is messing the USB up in the first place???

BTW, what is the voltage on -FLAG (U4 pin 4)? It should be 3.3V (once the 4V thing is sorted).

Edit: separate thing... what is the tolerance of the resistor R25? It should be 1% according to the datasheet. I said in an earlier message that resistor tolerances of 5% are usually good enough unless otherwise stated. The datasheet for the FE2.1 clearly states otherwise.

Edit 2: this is from the FE2.1 datasheet. Hopefully the chip is ok once you sort the 4V problem. Other 3.3V chips on the board will likely have similar absolute maximum ratings.

MaxRating.jpg

[DeKay]

Problem seems to be that all of the 3.3V lines are showing as 4V. XIN is 3.3V but XOUT is 4V. So... the crystal is bad?

If the Crystal was bad but a signal is getting from XOUT to XIN, wouldn’t that just mean the timing of the controller would be off? Or does a wrong voltage at XIN stop it working completely?

The crystal oscillates at a single frequency and can't be tuned by adjusting the voltage around. I'm not sure what the XOUT voltage would be without knowing the internals of the FE2.1. It is not as simple as volts in -> frequency out. LeeW needs to sort the 4V problem out first.

I mean... if you look at the datasheets for the two crystals on the Mister Wiki, neither even mentions voltage. What they do mention is their frequency is accurate to around 30 parts per million. I think you'd need a scope to verify that the crystal was operating properly.

[DeKay]

Does anyone know how long the internal Ethernet and USB cables that connect to the DE-10 are supposed to be? The multisystem manual just says the included cables are “short”.

Based on the pictures on page 12 and 13 of the manual, and knowing the dimensions of the Nano and Multisystem boards, you should be able to determine the length pretty accurately. But to make a little more space, see what others are doing on the last page of this thread. Or... just ask somebody in that thread to measure theirs if you really want to be precise about it.

viewtopic.php?t=3031&start=330

[emiliom]

Does anyone know how long the internal Ethernet and USB cables that connect to the DE-10 are supposed to be? The multisystem manual just says the included cables are “short”.

Based on the pictures on page 12 and 13 of the manual, and knowing the dimensions of the Nano and Multisystem boards, you should be able to determine the length pretty accurately. But to make a little more space, see what others are doing on the last page of this thread. Or... just ask somebody in that thread to measure theirs if you really want to be precise about it.

viewtopic.php?t=3031&start=330

I did do a bit of measuring and I recon they’re probably 15cm. I couldn’t find shorter than that on Amazon anyway. The DIY Adafruit cable looks interesting. Certainly doesn’t look like it would put any pressure on the Ethernet port.

[DeKay]

I did do a bit of measuring and I recon they’re probably 15cm. I couldn’t find shorter than that on Amazon anyway. The DIY Adafruit cable looks interesting. Certainly doesn’t look like it would put any pressure on the Ethernet port.

All those cables twisting around in there certainly makes for a tight fit. If I were building one up, I'd also consider crimping my own "short" Ethernet jumper cable so I could avoid the strain reliefs from bought cables. The strain reliefs are more a negative than a positive in something like this where the cable will never move once the assembly is done.

[LeeW]

Not going to lie, the usb and ethernet cable situation in this area of the board is frankly terrible. If your USB or ethernet adaptors are too long or too fat it won't fit (which is actually most cables).

You have maybe a maximum of 2mm clearance from the edge of the USB A male connector:

2o6MfSg.jpg

From below:

Y3RUMdv.jpg

And about 22mm clearance for the micro USB side:

dbHhYjP.jpg

And the USB A port is directly in front of the DE10 board ethernet connection:

LjRiw5e.jpg

So if you have large grips on the Ethernet cable it gets in the way (the plastic part is touching, maybe a bit difficult to tell because it's transparent):

o7Vokjt.jpg

It can squeeze in but there's tension from the springy bit:

8uBjBrQ.png

EDIT: I would not be comfortable keeping it like this and will probably cut these clips off the cable.

I'm going to try measuring the voltages and things again later with a different multimeter. The one I was using was an old beat up one with a bit of a dodgy probe. So I'm hoping I can get better measurements and also measure the crystal frequency.

[emiliom]

Not going to lie, the usb and ethernet cable situation in this area of the board is frankly terrible. If your USB or ethernet adaptors are too long or too fat it won't fit (which is actually most cables).

Agreed!

You have maybe a maximum of 2mm clearance from the edge of the USB A male connector:
From below:

And about 22mm clearance for the micro USB side:
And the USB A port is directly in front of the DE10 board ethernet connection:

So if you have large grips on the Ethernet cable it gets in the way (the plastic part is touching, maybe a bit difficult to tell because it's transparent):

It can squeeze in but there's tension from the springy bit:

EDIT: I would not be comfortable keeping it like this and will probably cut these clips off the cable.

I had an ethernet cable arrive from Amazon today which I bought last night and thought would be fine...and the connectors don't fit. The cable is perfect, but unless you have a perfectly straight slim connector it's just not going to fit. What's annoying is that I do have cables that will fit perfectly at home, but they're 1-2m long. And I have a USB A - Micro USB cable that also fits perfectly, but it's 1.8m long! I can use those for testing but I'm not sure which way I'll go yet with the permanent cables.

[rezendes]

I just estimated the length of the Ethernet/usb to be around 6”, I have a spare micro usb that I’ll be using once I get the last component (FE2.1 IC) and also I’ll share the Amazon link to the Ethernet cable I bought if it turns out working well. Also I kind of had to take a chance on the crystal oscillator because I couldn’t find the exact one listed on the BOM… we’ll see if everything works or not and I’ll come back here to report. Oh, also got my Toslink optical connector ($14) after shipping for one component and I’ll probably not ever use it but whatever I just didn’t like the empty spot on the back of the board…. As soon as I soldered it on and plugged in the HDMI (without DC power) it lit up the red light of the jack so that’s a good indicator I think? I still need to somehow confirm the mini expansion type slot connector up front is functioning properly, I don’t have a mt32 pi or whatever it’s called.

[emiliom]

Ok, so I finally tested the setup today, and I have something working but to be honest I'm pretty confused.

Board_Connected.JPG

Connecting it up to the HDMI input on my monitor I get this:

Monitor_Output.JPG

I thought my monitor would autoset, but the frequency specs for my monitor are Horizontal: 30-83 kHz, Vertical: 56-75 Hz, so 30Hz for vertical is certainly out of range. Can I configure the system to output a frequency that is within range?
I thought I'd try it on a TV just to check, and success! I got the initial splash screen, then the DE-10 booting, then this:

TV_Output.JPG

Problem is I couldn't select anything because I had no input devices connected. So I turned the MSS off, plugged in a keyboard and mouse and rebooted. Booted all good again, but no keyboard or mouse function.
I had already checked continuity from the USB ports to the FE2.1, and I checked the voltages on the chip pins and they are all ~ 3.3V. I did few few other random voltage checks around the board and it seems I'm getting 5V and 3.3V where I should, so that looks ok.

Have I just jumped into this without setting up the DE-10 first? Am I supposed to be connecting it to my PC and configuring it and then plugging it into my Multisystem? Or am I just missing something obvious?

[rezendes]

I used Mister Fusion and Update All, don’t think I did anything specific to the de10 nano, followed this video

[LeeW]

Have I just jumped into this without setting up the DE-10 first? Am I supposed to be connecting it to my PC and configuring it and then plugging it into my Multisystem? Or am I just missing something obvious?

Yeah you do need to set up the MiSTer FPGA, you need to run mr fusion first: https://github.com/MiSTer-devel/mr-fusion

What I haven't really seen mentioned too much is that windows does not seem to like one of the partitions on the USB after going through this process and this will always pop up on windows with an error when trying to read the SD card when first connected. So drive E may show an error, but the actual data will be on drive F or something.

After the SD card is set up you'll also need to then download the cores and stuff, you can do this manually but the Update All script will automate this: https://github.com/theypsilon/Update_All_MiSTer

[rezendes]

Just got the USB IC literally right after my last post. Popped it in and everything seems to be working fine, all USB are Go! Worth noting that the short _6” USB cable I planned to use didn’t work because I assume it was a Power Only cable with no Data (Hate that those exist). I’ll have to trim this other USB down quite a bit and the Very last thing I’m waiting for is the Ethernet cable. I already contacted my local 3D print man and he said he’ll get back to me in a couple days with a quote regarding the standard mister multisystem case and buttons. Perfecting my retronas setup right now as well. Getting closer!

[LeeW]

Reading through the data sheets I'm now confident that my USB issue is the power IC (U4), when inputting the BOM I must have accidentally selected the wrong component. The USB controller appears to be active low on USB_EN while my power IC is active high. 1 digit wrong on my part ID.

Other than that, I think my board is probably fine, I don't think the 4V readings I was getting were accurate after testing again today with a better multimeter.

[rezendes]

@multisystem I was wondering if there are any plans to release the PCB files for the Pi-MMS as well?

[emiliom]

I used Mister Fusion and Update All, don’t think I did anything specific to the de10 nano, followed this video

Yeah you do need to set up the MiSTer FPGA, you need to run mr fusion first: https://github.com/MiSTer-devel/mr-fusion

What I haven't really seen mentioned too much is that windows does not seem to like one of the partitions on the USB after going through this process and this will always pop up on windows with an error when trying to read the SD card when first connected. So drive E may show an error, but the actual data will be on drive F or something.

After the SD card is set up you'll also need to then download the cores and stuff, you can do this manually but the Update All script will automate this: https://github.com/theypsilon/Update_All_MiSTer

Done that now, although my monitor still didn't want to know. Seems that the MiSTer is trying to auto detect my monitor but failing - it kept trying to output 30Hz vertical. I fixed this in the MiSTer.ini file so now it has the correct resolution and vertical frequency and I can see it ok.

But I still seem to have no USB. I have continuity along the data lines between the USB ports and the FE2.1, but when the system is on I have zero volts at every connection on every USB port apart from JP11 where I get 1.25V and 3.3V on pins 2 & 3. I get continuity with GND on pins 1 & 4 for every USB port except JP11 which, if I'm reading the diagrams correctly, would seem to make sense as VBUS connects to GND via capacitors, except for JP11 where VCC connects to nothing. I'm at a bit of a loss.

EDIT: With the MiSTer connected I get 4.87V on VCC on JP11 and zero on the other 3 pins.

[DeKay]

But I still seem to have no USB. I have continuity along the data lines between the USB ports and the FE2.1, but when the system is on I have zero volts at every connection on every USB port apart from JP11 where I get 1.25V and 3.3V on pins 2 & 3. I get continuity with GND on pins 1 & 4 for every USB port except JP11 which, if I'm reading the diagrams correctly, would seem to make sense as VBUS connects to GND via capacitors, except for JP11 where VCC connects to nothing. I'm at a bit of a loss.

So I think you are saying that VBUS, U4 Pin 6 reads 0 Volts when everything is powered up? What are the voltage for FLAG- and USB_EN on that chip?

• If -FLAG (an output from U4) is around 0V, U4 detects a problem like over current (i.e. a short somewhere)

• If USB_EN (an input to U4 from the FE2.1) is around 5V, then FE2.1 isn't happy for some reason

Note that the datasheet for U4 is very confusing. It talks about an EN (active high) and _EN (active low) signal but the drawings for the chip in the schematics only show EN and never _EN. However, Table 6 Ordering Information sorts this out. The NCP380LSNAJAAT1G shown in the schematic has "Active Enable Level Low", so it is _EN. I suspect this is where LeeW made his mistake. AFAICT NCP380L... part numbers are Active Low and NCP380H... part numbers are Active High.

Active Low (_EN and NCP380L) in this design makes sense. The Power Enable PWRJ pin on the FE2.1 is Active Low. The USB_EN line is pulled up (deactivate U4) by R22 with a 10K resistor as a precaution to shut off U4 unless everything is working.

So what does all this mean?

• Your U4 better have a part number of NCP380L...

• the schematic symbol for U4 should be cleaned up. Pin 3 should show as -EN (active low) and not EN (active high)

• the net USB_EN should also show as -USB_EN to indicate active low

A comment I made earlier in this thread that USB_EN should be 5V was incorrect for this reason.

Other things to check if your U4 part number is correct

• verify R20 is propery connected

• verify C29 is propery connected

Now having written all that, I looked again at what you said. "I get continuity with GND on pins 1 & 4 for every USB port... would seem to make sense as VBUS connects to GND via capacitors"

This doesn't make sense. Capacitors don't provide continuity, they provide storage of electrical charge. The schematic symbol of two parallel lines that don't touch each other emphasizes this fact. If they did provide continuity, then all of those decoupling capacitors (e.g. C30 - C35) would be short circuiting the power supply.

If, with everything powered off, you get some very low number of ohms between VBUS and Ground, you have a short circuit there somewhere. U4 is sensing that and shutting VBus down. Find that short and you're golden. If you are getting something well into the kiloOhms or greater, that is various parasitics and you're probably OK.

[emiliom]

But I still seem to have no USB. I have continuity along the data lines between the USB ports and the FE2.1, but when the system is on I have zero volts at every connection on every USB port apart from JP11 where I get 1.25V and 3.3V on pins 2 & 3. I get continuity with GND on pins 1 & 4 for every USB port except JP11 which, if I'm reading the diagrams correctly, would seem to make sense as VBUS connects to GND via capacitors, except for JP11 where VCC connects to nothing. I'm at a bit of a loss.

So I think you are saying that VBUS, U4 Pin 6 reads 0 Volts when everything is powered up?

Very nearly. Looking into it a bit more I realised the voltages are fluctuating. I'm guessing this is the crystal oscillation? I get 0.06 - 0.26V, so almost nothing.

What are the voltage for FLAG- and USB_EN on that chip?

• If -FLAG (an output from U4) is around 0V, U4 detects a problem like over current (i.e. a short somewhere)

• If USB_EN (an input to U4 from the FE2.1) is around 5V, then FE2.1 isn't happy for some reason

FLAG-: 3.27 - 3.30V
USB_EN: 4.33 - 4.56V

• Your U4 better have a part number of NCP380L...

Yes, I have an Active Low part: NCP380LSNAJAAT1G

Other things to check if your U4 part number is correct

• verify R20 is propery connected

• verify C29 is propery connected

They both look properly connected. I re-soldered them just in case, but no difference.

If, with everything powered off, you get some very low number of ohms between VBUS and Ground, you have a short circuit there somewhere. U4 is sensing that and shutting VBus down. Find that short and you're golden. If you are getting something well into the kiloOhms or greater, that is various parasitics and you're probably OK.

It had been looking suspiciously like a short. I was hoping there was some reason that I didn't understand that the VCC USB pins were showing continuity with GND, but obviously not. Resistance between each VCC pin and GND for USB1-7 is <1 Ohm. Voltages when powered on oscillate and range from 0.01-0.05V and 0.02-0.11V. So a very, very tiny amount of current is getting through, but most is leaking away.

JP11 (or USB0?) seems ok. VCC is 4.85V. So it looks like the short is somewhere after JP11. Would that be a reasonable assumption?