5m USB Too Long for Some Devices?

[fatwoul]

I know the "limit" with USB is supposed to be 5m, but are some devices more or less happy with this? I am specifically talking about my Atik Titan, which is becoming one of the most tiresome objects in my observatory.

If I plug it into either of the two 5m A-B USB from my desktop to my pier, it glitches and freeze, whether or not the camera itself is powered (with the Titan, powering is optional). If I plug the camera into the provided 3m USB, it usually functions correctly, although does so more reliably if the camera isn't powered.

My 314L+, which of course has to be powered, has never presented these sorts of problems, which leads me to wonder if it's down the Titan's need for power down the USB. I seem to recall the explanation from Atik is that the Titan only requires power for the cooling fan, but if the camera can't function on a 5m when powered cameras can, it seems a bit duff that the optional power can't/isn't used to also power the USB, which I assume the 314L+ does.

I've got a powered hub, which I am going to try with the Titan. If that fixes the problem, then I'll have to work out a way to get 5V to the pier (for the hub), as well as all the 12V already going there. I am definitely putting another tick in the "cons" column of the Atik Titan.

[msinclairinork]

I think we've all encountered or are encountering USB issues at some point! (you're not alone!)

IMO there's no book on it...each device/cable/hub/PC etc pairs up differently..why? - I don't know.

All I can say is trial and error seems to be the only way. Keep trying different things and eventually you'll find a method that works for you.

Good luck!

Michael

[fatwoul]

Thanks Michael. I have to say, this revelation is a mixed blessing - it's a good thing in that I have finally found out why my imaging rig was so twitchy, but it's a bad thing because it means I may have to complicate my setup with powered hubs.

The biggest problem with powered hubs, for me, is that my dome desktop REALLY doesn't like having both the Atiks plugged in at once. I've always run the Titan through my laptop and the 314 through the desktop, but when I tried running them with the same machine the poor thing just got into a tizzy.

I'd love to think that a powered hub would solve that too, as the USBs trailing everywhere would be so much simpler if they shared the main leg of their journey, but somehow I think I'm being too optimistic!

Anyway, for once I'm thankful for a few cloudy days and nights, so I can sort this out.

[johninderby]

Have you tried an active repeater USB cable?

You can readily get 5, 10 or 15 metre ones quite cheaply. I had a camera that worked fine with a 3m cable but just wouldn't work properly with a regular 5m cable but now works perfectly with a 10m active cable. Mine doesn't need an external power supply for the cable, it just plugs in in place of a regular USB cable.

John

[pete_l]

I know the "limit" with USB is supposed to be 5m, but are some devices more or less happy with this?

Yes, without a doubt. Even if devices are able to send their signals over a 5 metre cable, they may not be able to power themselves from it.

I discovered this the hard way a few years ago. I had a webcam that worked perfectly "on the bench" when connected to a nearby computer. When I took it out to the telescope and ran a long cable to it, the PC simply didn't recognise there was anything connected. After a lot of futzing about, the problem turned out to be the current the webcam drew was causing a large voltage drop over the length of the cable. What started out as +5 Volts at the PC end of the cable was barely 4 Volts at the camera. Powering the webcam externally made everything better, but taught me a lesson.

[Parus major]

I dikd at one stage have a 5m active USB cable which was bought for the great sum of £6 running from an indoor desktop to a Belkin powered USB hub which in turn was connect to my telescope mount, Philips webcam and the hard to comprehend combination of a USB to Playstation 2 controller converter with a Wireless dongle for it. It all worked fine but if you disconnected the cable from the pc even without the devices on the other end it would bluescreen the system. Windows wouldn't boot if you left it in but would when you took it out. As ticks go those are easy to live with.

[Gina]

Yes, without a doubt. Even if devices are able to send their signals over a 5 metre cable, they may not be able to power themselves from it.

I discovered this the hard way a few years ago. I had a webcam that worked perfectly "on the bench" when connected to a nearby computer. When I took it out to the telescope and ran a long cable to it, the PC simply didn't recognise there was anything connected. After a lot of futzing about, the problem turned out to be the current the webcam drew was causing a large voltage drop over the length of the cable. What started out as +5 Volts at the PC end of the cable was barely 4 Volts at the camera. Powering the webcam externally made everything better, but taught me a lesson.

Yes, I've found the same. The volt drop is the limiting factor on USB cable length. I used a powered hub at the far end of the cable from the computer.

[radiostationx]

There are some usb extension cables with a small box of electronics attached in line.

This helps with the drop in the 5v supply voltage.

I have one unused .

[johninderby]

I dikd at one stage have a 5m active USB cable which was bought for the great sum of £6 running from an indoor desktop to a Belkin powered USB hub which in turn was connect to my telescope mount, Philips webcam and the hard to comprehend combination of a USB to Playstation 2 controller converter with a Wireless dongle for it. It all worked fine but if you disconnected the cable from the pc even without the devices on the other end it would bluescreen the system. Windows wouldn't boot if you left it in but would when you took it out. As ticks go those are easy to live with.

I would suspect a dodgy active usb cable. Is it one of those eBay specials? The one I've been using works just like a regular usb cable as far as plugging and unplugging it. The electronics are fitted inside the usb plug so it looks just like a regular usb cable.

Also a good cable usually won't have as much of a voltage drop as the cheaper cables which can make all the difference.

John

[Sp@ce_d]

I use this system USB 2.0 Active Extension System Pro - Extend USB 2.0 up to 56m | LINDY UK and have standard 5m cables running from the active hub. You can get quite a long run.

The other thing to consider is total bandwidth and total current draw requirements of all the attached devices. If they exceed what's available from the USB interface then that's the bottleneck. In that case you'll need to distribute across separate controllers. i.e. you can have multiple USB ports (connectors) on the PC but they are sharing the same controller. In that case a separate card installed into your PC would probably be required.

[AstroTux]

One thing you my need to check is the USB card in your desktop. There are a number of 4 port USB 2.0 cards on the market and you would think that the USB rule of 500mA per port would apply to each of them, but in reality it does not!!!!

The rules is 500mA per HUB, and a number of 4 port cards actually contain two hubs with 2 ports per hub. If you are drawing current from a device attached to one port of the hub you will not be able to get the full 500mA from the other port.

My advice would be to check that you are not using any other USB devices attached to the same card.

[Parus major]

I would suspect a dodgy active usb cable. Is it one of those eBay specials? The one I've been using works just like a regular usb cable as far as plugging and unplugging it. The electronics are fitted inside the usb plug so it looks just like a regular usb cable.

Also a good cable usually won't have as much of a voltage drop as the cheaper cables which can make all the difference.

John

Comet, actually. The thing about tech is that you can keep buying until you find one that does the job. This one does the job with a couple of odd ticks that don't go on while it's in operation so I'm happy with it. It took me a while to find one that actually operated correctly with what was on the end of it so I'm really not in the mood to start looking again.

[Parus major]

One thing you my need to check is the USB card in your desktop. There are a number of 4 port USB 2.0 cards on the market and you would think that the USB rule of 500mA per port would apply to each of them, but in reality it does not!!!!

The rules is 500mA per HUB, and a number of 4 port cards actually contain two hubs with 2 ports per hub. If you are drawing current from a device attached to one port of the hub you will not be able to get the full 500mA from the other port.

My advice would be to check that you are not using any other USB devices attached to the same card.

Does that apply to the straight off the motherboard types with the single USB and the cable on the back that fit in the expansion bays? I've got one on my build in the vertical bay for fan controllers.

[johninderby]

Comet, actually. The thing about tech is that you can keep buying until you find one that does the job. This one does the job with a couple of odd ticks that don't go on while it's in operation so I'm happy with it. It took me a while to find one that actually operated correctly with what was on the end of it so I'm really not in the mood to start looking again.

You should see my "bit & pieces drawer" that's full of old cables, adaptors etc. that almost work.

John

[AstroTux]

Does that apply to the straight off the motherboard types with the single USB and the cable on the back that fit in the expansion bays? I've got one on my build in the vertical bay for fan controllers.

I'm not sure about these as I have no experience with them, but if they are powered from a source other than the USB motherboard header then they may provide the 500mA per hub you require. If they only take their power from the USB header then the entire device will be limited to 500mA.

As a quick note on hubs, un-powered hubs can only provide 100mA per port, so be careful with your connections.

[AstroTux]

Just had a re-read of the USB 2.0 specification and thought you might want a 'heads-up' on what it says (I have paraphrased it a bit so as not to confuse you with some of its wording) ......

"• Root port hubs: Are directly attached to the USB Host Controller (i.e. they are part of the USB card in your PC). Hub power is derived from the same source as the Host Controller (i.e. the PC motherboard) and they MUST supply at least 500mA to each port. Such ports are called high-power ports. Ports that can supply only 100mA are termed low-power ports.

• Bus-powered hubs ( i.e. the standard hub with a cable that attaches to your PC USB port): Draw all of their power for any internal functions and downstream facing ports from the upstream facing port. Bus-powered hubs may only draw up to 100mA upon power-up and 500mA after configuration. External ports in a bus-powered hub can supply only 100mA per port regardless of the current draw on the other ports of that hub.

• Self-powered hubs ( i.e hubs with an external power supply ): Power for the internal functions and downstream facing ports does not come from the upstream facing port. However, the USB interface of the hub may draw up to 100mA from its upstream facing port to allow the interface to function when the remainder of the hub is powered down. Hubs that obtain operating power externally (from the USB) must supply 500mA to each port.

• Low-power bus-powered devices: All power to these devices comes from VBUS ( the +5V line of the USB cable). They may draw no more than 100mA at any time.

• High-power bus-powered devices (most webcams, etc fit into this category): All power to these devices comes from VBUS. They must draw no more than 100mA upon power-up and may draw up to 500mA after being configured.

• Self-powered functions ( e.g. printers, anything with an external power supply): May draw up to 100mA from VBUS to allow the USB interface to function when the unit is powered down. All other power comes from an external (to the USB) source."

This tends to indicate that all ports on a PC USB card should provide at least 500mA, but I have discovered that quite a few cards tend to ignore this part of the specification (makes you wonder why it was conceived doesn't it ).

Interestingly enough, the specification does not make any mention of maximum cable length but does go into a detailed examination of the maximum permissible voltage drops to devices, and this will depend upon a number of factors including the quality of the cable.

It also references voltage 'skew'. This is an interesting phenomenon where the two signals arrive slightly out of phase with each other (USB uses a + and - signalling system). This could also be caused by a number of things such a cable quality, shielding against other stray signals (motor controllers can kick out quite high levels of electro-magnetic interference), cable length, etc.

All in all, the spec doesn't really help does it?? However, I would be careful with the quality of cabling if it being used anywhere near a motor.

[Coco]

When I first started imaging 4 years ago with a 40D (+ astrotrac) I used to use a 1.5m usb + 10m usb extension without any problems, since moving onto cooled CCD and mounts etc it all went down hill fast

I think the remark made earlier is best... no rules per device.. If its a long way power it up, but expect an issue on one of those clear nights just when you dont want it

[tibbs1972]

How about a wireless USB Hub local to the telescope, connected to your internet router, around £25, if you want to get more extravagant.