DS-3 coax strange problem

[Dialtone]

Hope someone can help on this one.
I have some telecom equipment that interconnects a DS-3 signal (~45 MBS) that is having trouble syncing up if I use a coax of around 20 foot length.

The cable tests OK and even works fine in other pieces of equipment. It even works on the suspect equipment, if I loop the signal back to iteslf thru the same coax (effectively doubling the length of the conductor path)

I did some quick and dirty calculations and found out that the wavelength of a 45MHZ signal is approx 21.6 feet.

Is it possible (and probable) that the trouble I am seeing is related to the frequency and length of the coax involved?

I am really stumped to explain this one.

Thanks
Dialtone

 

[Nigel Goodwin]

I've no idea about your specific problem?, but it's always been known that specific lengths of cable on RF circuits can cause problems on specific frequencies - usual cure is to cut a few inches off the lead.

 

[JimB]

If a co-ax cable is not terminated in its characteristic impedance, you will get peaks and nulls in the voltage as you go along the length of the cable.
I have never heared of DS-3 before,
but I would expect that either:
In the equipment, there is a means to terminate the cable by switching in a suitable resistor.
or
You should use a T-piece and fit a terminator external to the equipment.

As I have been typing, another thought occurs to me, are you using the correct cable?
Co-ax is made with 50, 75 and sometimes 91 ohm characteristic impedance. Even when correctly terminated, using the wrong cable will give you some strange effects.
And as Nigel said, the effects will vary with the length of the cable.

Edit: I have just been Googling and found that DS3 uses 75ohm cable. If your cable is not 75ohm you could easily get problems.

JimB

 

[Dialtone]

The coax is the correct type for the application (AT&T type 735).
The equipment has options to set the signal attenuation for the length involved, and has been set accordingly.
I should also mention that the signal is not an RF signal, but a digital bitstream of approximately 45 Mega-bits/second. DS-3 is used in North America to refer to one standard MUXed telecom signal format. Other parts of the world use a similar but slightly different standard that I think is referred to as E3.

 

[audioguru]

When I worked with digital telephones on twisted pair cable, some didn't work when the cable was a certain length. Changing the cable's length fixed them.

 

[Roff]

The coax is the correct type for the application (AT&T type 735).
The equipment has options to set the signal attenuation for the length involved, and has been set accordingly.
I should also mention that the signal is not an RF signal, but a digital bitstream of approximately 45 Mega-bits/second. DS-3 is used in North America to refer to one standard MUXed telecom signal format. Other parts of the world use a similar but slightly different standard that I think is referred to as E3.

You seem to have ignored JimB's comments: It must be properly terminated (in approximately 75 ohms), or you will get nulls at certain frequencies. Source termination OR load termination alone will work IF the termination is really 75 ohms, with little parasitic capacitance. Source AND load termination is much more robust.

 

[gerty]

Had a simillar problem when I tried to clean up a network rack by making up some shorter jumper cables. The original cables were about 8' long and I shortend them to about 12". The whole rack went down , one of the oher techs informed me that they had to be at least 24" long ..Took me a while to live that one down.

edit: the above lengths may not be relevent to your situation, but short cable lengths can cause problems. they were terminated
50Ω