ADuM4160 PIN pin grounding
Q. I am getting “USB device not recognized” error message – what do I do? Also, since the isolator is soldered into my circuit, “reconnect peripheral” suggestion seems too difficult to follow.
I decided to write this article after receiving several e-mails from people who bought my isolator. While setups described in those e-mails were different, the problem was the same – a PC refusing to recognize the device connected through the isolator. Here I will try to explain what is happening and also share my ideas how to troubleshoot and possibly fix the problem.
When nothing is connected to USB port, the bus is held at ground level with pull down resistors on the host side. USB device, when connected, pulls one of bus lines up, often times also with a resistor connected to Vbus and data line. Host sees it, sends bus reset and tries to query the device. If device is answering, host keeps querying the device and eventually enumerates it. When device is enumerated, application takes over.
If device is not answering (like for example, when self-powered device is turned off), host will give up and post “Device not recognized” message. To get attention from the host, we need to generate bus event, i.e., disconnect the peripheral and connect it back again.
If we are dealing with typical copper-wire-connected USB host and peripheral, which end of USB cable gets disconnected and reconnected would not matter. However, when host and peripheral are talking through ADuM4160, this is not always the case. I tested about 50 different USB devices and found out that while breaking the connection on the peripheral side works every time, host side disconnect does or does not help depending on device.
It is possible to simulate device disconnect without breaking actual connection by pulling PIN pin of ADuM4160 to ground. Normally it is held at 5V by R8 pull up resistor. This resistor is located next to peripheral side USB connector and marked 1M above it. Left side of it, the one right below the “1M” mark is connected to PIN pin (12). The closest ground point is right pad of bypass capacitor, which sits above R8. An experimenter with good eyesight and steady hand can easily connect those two points with tweezers. However, if said experimenter is not feeling comfortable operating tweezers inside powered circuit, a safer arrangement is shown on the title picture.
Disconnecting peripheral, either directly or by driving PIN pin, with power applied to the isolator and host side connected, will most likely solve device detection problem. If you still having issues with the circuit, please let me know, I will try to help.
Oleg.
Related posts:
Dear Oleg,
Your PIN advice works great. I installed a toggle switch between ground and R8 that allows me for “re-boot” the connection in case of accidental disconnection. I use your isolator on my USB audio DACs. I strongly recommend it for this purpose as the improvement in sound quality is evident.
Many thanks fro your help.
PS: a remark; I read 3.3VDC at R8 (not 5VDC) as it is connected to VDD2…
Thank you for the catch – I’ll make corrections to the article!
hi Oleg,
the board works great with a mice in low speed mode
in fast speed mode i get only the USB device not recognized error.
I try to use an usbasp Adapter , no way, not slow or fast.
I get only USB device not recognized.
regards from Germany
Uwe
From what I can see, USBasp should be compatible with this isolator. Have you tried other high-speed devices, like usb flash drive, for example?
Also, tell me about your power supply – can you measure 5volts with USBasp connected?
Hi Oleg,
I have power off my Laptop, not only a reboot.
Now i works well, in fast mode and with usbasp!
A good job….
regards from Germany
Uwe
I’m glad you got it running; however, powering the isolator from a computer is wrong. It might work but it won’t isolate. You need to use separate power supply fro peripheral side.
Hi,
Have you tested your USB Isolator with external 2.5inch portal hard drives?
Thanks
I haven’t tested it with 2.5 drives. It works fine with 3.5 external drives. What is the difference?
Hi,
I made my own circuit using this device, but it works fine on some PC’s, but on other PC’s it says, “USB device not recognized”. What do you think could be the problem?
I have pins 3,4,5 connected together and then on the other side, 12,13,14 are connected together. I am also using 24R on both sides for the D+ and D- lines.
Best Regards,
Ray
I am not using 1meg pullups or any pullups, just connecting them directly.
The circuit works on one PC, but on others there is the not recognized problem.
Have you tried grounding PIN pin as in the article (to do this, you will need a pullup on it)?
Hello and Good day
My problem is this I am using a 4160 ADUM typical configuration with the proposed Analog Divic, but when you first connect a device such as a keyboard or a mouse everything works fine. The problem is when I unplug the mouse or keyboard and put a USB flash memory tells me unrecognized disposition
The same happens if other circuits attached for the first time a USB flash memory everything works fine but when I disconnect to connect a mouse or keyboard when I connect it tells me unrecognized device that could be happening
Thanks
Hello,
I have now built and tested this device, but am experiencing strange results. Do not have issues with connected devices being recognized, but no matter what (full-speed) USB device is connected, the host computer reports that “this device can perform faster” — even though the isolator is plugged into a full-speed host port. If I simply eliminate the isolator, plugging the USB device directly into the host port, all is normal. Any suggestions?
Cheers
In USB 2.0, 3 speeds are possible – low, full, and high. You are using high speed host and device, isolator supports low and full speeds only. As long as your device is working, you can ignore this message.
Follow-up… after much swap’n and test’n it is now confirmed that the USB drivers for the Musiland Monitor01-USD conflict with whatever the ADuM4160 invokes when connected to the host computer. This holds true for XP and Win7x64, regardless of what USB device is connected to the isolator. For example, an external Seagate USB HD normally performs at ~32MB/s but if connected via the isolator it then is only able to muster ~1MB/s – IF the Musiland driver is installed. Remove that driver and all works as otherwise expected. This behavior is consistent with the “device can perform faster” notices mentioned in the last post… everything works, just at USB-1 speeds
I will certainly contact Musiland about this, but if anyone here has further information that would be nice to have too.
Cheers
The circuit is ok, but even I set it to ful speed, a windows alert says: “it’s not a full speed peripherical” with the isolator mounted and powered with an external supply. Do you knowe how to solve this problem?
I haven’t seen this error before. Can you make a screenshot of error message and send it to me to mazurov at circuitsathome com?
Hi,
I just connected my two boards today for the first time. I’m using Windows XP with a USB 2.0 cable, testing with a USB memory stick, and have both jumpers in the lower “Fspeed” positions. When using both boards, I get a message at the lower right of my screen saying “This device can perform faster. This USB device can perform faster if you connect it to a Hi-Speed USB 2.0 port. For a list of available ports, click here.” When I plug the memory stick into the port directly, the message does not appear.
-Bryan
This message means that Windows was not able to negotiate high-speed with high-speed device, which is not surprising since ADuM4160 supports full and low speeds only. Your drive should work just fine though; that’s exactly how I check isolators prior to shipping – I connect USB drive with large zip file to a PC via an isolator and run archive integrity test.
But it’s USB 2.0 full speed full compatible?
Hallo Oleg
I can”t get my usb isolator to work, PC/W7/64.
I was recommended by a DIY group build for audio, the usb entry
pcb work’s fine solo. The aim is to isolated pc-ground.
I cant do a desent test!!
Groet Ton
Hi Here some more info.
I did the wire at the chip , the pc react,wire away return to
where it was. i powered with a wirebridge to the vin points.
The USB pcb (= usb to I2 s interface) by
firm name altmustech” from japan/china.
It works stand alone, W7 reacts direct when cables switched.
Groet Anton (61) Holland
Didi you check the isolator with some other USB device, like, for example, USB flash drive?
Hallo Oleg
I trieed with common-usb sticks, give’s no reconnicion sign from windows. i get the feel it wont come over the chip.
Groet Ton
Hello Oleg
After some more testing , i find out (a) that if i losen the usb cable to and from isolator and usb interface windows react accordingly. (b) usb interface has a DSP with a Vbus 5volt!!
As eyesided is not ok and age (61) is making this may be a bridge to far!!
Groet Ton
Hi Oleg,
Like Igo51, I get the “this device can perform faster” message. Tried three different PCs running XP and same on each. It reads memory sticks OK, but I bought mine for use with a recording device (E-MU 0204) that needs usb 2 speeds to work right. At the moment, it sees the usb link as being too low in bandwidth to allow the high sample rates I want. Do you have any further solutions?
Thanks,
Hugh
The isolator does not support high speed, only full and low. I’m not aware of any high-speed capable USB isolators, sorry.
Good morning, Oleg!
Why do not you put on the slave power galvanically isolated DC-DC converter?
What do you mean – I do have DC-DC converter on the slave and it’s isolated?
I mean, isolated DC-DC powered from the host
I’m planning to have this as an option in the next version of the isolator. The issue is max.current – currently, all isolated DC-DC modules are rated at 100ma max. and I wanted to be able to provide more than that.
LTM8022 1A Isolated DCDC
Won’t work – it requires 7V Vin.
Excuse me, LT3439
At May 19, 2011 at 3:39 pm you wrote:
I’m currently prototyping low noise linear regulator for new “modular” USB isolator. The new isolator will be available with switching, linear, or USB-powered isolated supply. No ETA yet.
How do you do with this project?
Still working on it.
Sorry, but the specification defines the usb power consumption for USB 2 is 500 mA. Efficiency of modern DC-DC 72-90%. Total output at 400 mA should be ?
You are talking about current consumption. There is no specification which forbids USB Host port from sourcing more than 500ma. In fact, PC ports are often capable of 700+ ma and many hubs work in ganged current limiting mode, being able to source 2A from 4 ports – if you only use one port, you can have all of it.