USB Multiplexing

The MAX4999 is equivalent to a digitally controlled double pole 8 position selector switch. The binary combination on C0, C1, and C2 connects the COMI to one of the DataI lines and COMO to one of the DataO lines.

Are you driving the enable high?

Are you connected to the data lines which are selected by the channel select?

http://pdf1.alldatasheet.com/datasheet-pdf/view/233303/MAXIM/MAX4999.html

Thanks for taking your time to help!

Yes, I'm selecting the port directly by shorting the power (3.3V) to EN and/or select pins. I've tested the connections I've selected with the multimeter and, well.... it beeps!

I've been reading about the importance of the USB data lines being equal in length. Considering that I'm using a breakout board, where the traces are obviously different in length --> https://www.adafruit.com/products/1163, could that be the source of my problems?

Mati

I think you have the answer. 480MB/sec is fast and the wire length, material, location and shape (bends, curls, etc) will make a big difference.

Back in college (mid-late 80s') my friend an I took a senior level RF course. In the lab, our class had to design and build an AM radio receiver - they didn't tell us what we were building - each group received a part of the transmitter with input/output/amplification specs. Our part was the balanced mixer and we used a ring modulator with four diodes. On paper, it should've worked great, but due our lack of real life design skills, we never realized that the component leads and breadboard were causing all kinds of unwanted signals. One day, we had the spectrum analyzer connected to our mixer and out of nowhere we got a large unknown frequency pop up. We had no idea what it was from and our fear was that we somehow cooked one of the diodes - making it unbalanced. We started probing all around our mixer, then poof - the signal was gone! When we talked to our Lab TA, he asked us the frequency. He had us think about it, then he told us to go around to each group and ask what frequency they were working on. Sure enough, someone had turned on a signal generator and the long leads on our circuit acted like an antenna.

Case closed! But we never got our mixer to work and the class never got the AM receiver working.

Many too many) years ago I had a large computer from our company, go on the blink at irregular intervals....

Many tried to fix it, nobody could get it to run longer than 3 days at best......many parts were replaced, that worked in other systems of the same type without problems....

Finally, I was sent to stay there till it was fixed, or the US system designer would have to be sent for!!

I had my scope set up and a MW/FM radio for amusement.

Suddenly, it went wrong and the station I was listening to on MW went sort of mushy at the same time.

I saw some weird signals on the main voltage bus, signals I "sort of" recognised, but could not remember where or when......signals that should not have been there!!

(This is turning into a children's bedtime story!!!)

I was in the rest room next door, smoking a cigarette (so it was before 31-12-1976, which was when I stopped!), and staring out the window, with the door to the computer room open and the radio on loud, when I noticed an EW RADAR antenna revolving on the far horizon, some 10 miles away at a guess, in time with the noise from the radio!!!

It was a RADAR System that was causing all the problems.

The company had to screen their computer room in some way, but that was not my job, a special company did it if I remember correctly.....

I had seen some similar signals in the RN some years before!!

Cool story. Unwanted RF signals are all over the place and can wreak havoc on electronics.

I have another interesting story I'll post in a new thread in a couple minutes.

"I saw some weird signals on the main voltage bus, signals I "sort of" recognised, but could not remember where or when......signals that should not have been there!!

(This is turning into a children's bedtime story!!!)"

You tell ghost stories to children when tucking them in at night?

Oh, wait, Germany also has children's stories about mischievous elves that chop off the thumbs of thumb-suckers with giant scissors, and a story about a litter of kittens shoved into a sack and thrown into the sea to drown, the kittens dying one by one in horrible ways, until the last kitten finds a family to live with. (Never says they're a 'nice' family, or that they treat the kitten well. all we know is that they're offering shelter to the kitten, and we don't even know if it's conditional on being good mouser.) And then there's Krampus, who is busy hitting naughty children with birch switches while Santa is passing out presents to the good kids. Kramus also stuffs the REALLY naughty children into his wicker basket to carry them off " and they were never seen again."

Sometimes I forget that the land of my father's father's father has some really 'unusual' ideas about children's entertainment.

Bad hair day?

I hope you recover!!

So get well soon!!

PS. You obviously did not understand a word I wrote, or you would not have been so rude, or is that simply your general manner?

Note, nobody else felt the need to demonstrate a really bad upbringing...

But if you are fully prepared to learn something interesting, that is interesting for many technical people, I can expand it in a PM if you wish!!!

Or you ask simple questions and build up your knowledge..... Just to make it easier for you to understand.

But if you have no further interest, also not a problem for me!

Have a great day either way......

I believe we are having a problem due to cultural misunderstanding.

I was not attempting to be rude, and I apologize for any offense I have given.

I was trying to convey a tone of "That is shocking to me! But wait, when I look at the culture, it makes sense, so I should not be an 'ugly american' and assume that everyone tells the 'sugar-coated' stories american kids are raised on."

In my attempt to play the Self-Mocking Fool, I appear to have played the Boorish Oaf instead. I am sorry and ashamed.

Did I choose the wrong examples? Are those stories ones that are no longer considered acceptable? Or was it the 'tone of voice' I was writing in that turned my attempt at humor into an insult?

To me it was taken as a clever Dick being rude.....

Maybe another here thinks that I am over sensitive on the subject.

But it was a "fix", that helped make my "Tech Support" name when still working for Sperry Univac, very many, many years ago!! Not the only one I hasten to add....

Also, do not forget, I was born in the UK, not Germany!!

Just what I thought, over engineering the solution for a 12 USB port device. The Orico is a 13 port device. Did you consider that he may be superstitious and 13 isn't lucky?

Try this one, the true solution to a 12 USB port device: https://www.amazon.com/Satechi-Power-Adapter-Control-Switches/dp/B0051PGX2I/ref=pd_sim_sbs_147_6?ie=UTF8&dpID=51NSqsuiq%2BL&dpSrc=sims&preST=_AC_UL160_SR160%2C160_&psc=1&refRID=JRSVG73WKG360GH41EJP

Just kidding - I had no idea a 13 port device existed!

Thanks to everyone for replies!

@Rixter:

I will definitely read more about it, thanks for the links. I've found this website to be extremely useful as well: USB-in-a-nutshell.

1. So, do you think that a more appropriate PCB layout will most likely help my multiplexer problem?

2. I've been creating PCBs using an etching tank, is this a good enough process or should I consider having it outsourced and made professionally for the high-speed transmission lines not to cause reflections?

@tcmtech:

That's a great idea, I've been researching hubs and hub driving ICs for a while now, as it offers an even better functionality, than multiplexing. Circuits like this for example:

The reason I can't buy a ready-made hub as you suggested, is cause I'm building a mobile device, with very limited space and power available.

Answers to your questions 1 and 2 are both yes!

Crosstalks and noise isolation are real problems in pcb layout, print size / spacings, routing, etc. beside harmonics...are few that need to be addressed if you want reliabilty!

good luck!

One of the reasons that you're having trouble multiplexing your USB signal is that USB is not DESIGNED to be multiplexed.

Data on a USB line is grouped into 'frames.' If you're familiar with IEEE-232 signals (previously known as RS-232), you will see the similarities: both have 'start bits,' 'stop bits' and 'error checking.' but while RS-232 sends 7-8 bits of data between start and stop, USB has kilobytes of data between 'start' and 'stop.' That data is also bidirectional, one portion of the frame could be outgoing data being sent to a printer, followed by a section for the mouse to report its status, followed by a section of data going to a USB memory stick, followed by incoming data from another USB memory stick, etc. The size of each portion in a USB stream can vary, frame to frame, and the 'table of contents' for the frame is right after the start string and the clock reference, so all the USB devices 'downstream can see when it is their turn to pick off data or add in data. 'Breaking the line' as a multiplexer will do, causes the devices to see NO signal when a signal is expected, and they will interpret this as a 'frame fault,' and signal that the frame needs to be resent.

This is why we do not use 'multiplexers' to split USB signals, we use 'USB Hubs,' which take the entire signal and duplicate it to EVERY downstream port at once, and collect the signals from ALL the downstream ports, correct 'sub-bit' timing errors, and combine their signals into a single stream going back to the USB controller.

You put my thoughts, of which I was not completely sure, into words. Thanks.

I always think of a USB as being a "streaming" device.....correctly or not!!

First rule of troubleshooting: check your power. Make sure the voltages are correct. Need decoupling capacitors near the chips. Chips need a good ground. Make sure the decoupling capacitors are "good capacitors" at the frequencies and first few harmonics that you are generating.

Second rule of troubleshooting: carefully review your work with the first rule.