Topic: splitting transducer signal

[stuntman]

I was fortunate and found a 798c SI close out model for what they were getting cleared out for in the states (I am in Canada).

The feature that I would have wanted on the 2010 version is the Ethernet connection that allows 2 head units to share 1 transducer at the same time (Humminbird rep could not confirm this).  Everyone wants to see what is going on down there.

So what is the obstacle in splitting the transducer signal on the old units?

thanks,

sm

[RGecy]

Stuntman,

Welcome to the forum.  With the older units it is not possible to share the sonar signal between units at the same time.  You could use the US2W Unit Switch to switch two units between one transducer.  But only one unit at a time will be able to display the sonar data.

Hope this helps.

Robert

[stuntman]

I am looking for a technical reason why they cannot be split.
is it because the signal coming from the transducer amplitude based, not worth HB building or?

I know little about sonar and something about repeaters and cannot figure out why a splitter/repeater cannot be used or built.

Am I alone in wanting 2 units to work at the same time?

thanks,

stuntman

[RGecy]

Am I alone in wanting 2 units to work at the same time?

No not at all.  I am sure most of us do. 

Maybe Greg can step in and shed some light on exactly why this is.  If it could be done, I am sure HB would have done it by now.

Robert 

[FuzzyGrub]

I think the simple answer is that they can't because they don't have the H/W and S/W to support it. No ethernet, no high speed connection for one unit to share the data with the other.

Assuming you are talking about splicing the two units into one transducer cable, this at a minimum would cause severly distorted displays, and has the potential to blow the driver circuits in one or both units. 

WARNING: Assuptions made for the following on how HB units are designed, which I don't have privy too.

Think of each transducer element as a speaker and a microphone.  The unit pings by sending out the signal (speaker), and then waits for the reflected signal to return (microphone), and then interpret.   

When you tie both units to the one wire there is no way for them to synchronize.  One unit can transmit while the other is receiving.  The receiver doesn't know the signal is from another unit and just trys to processes it.  That is the simple case, but the send/receive cycle can overlap in time from completely out of phase to in-sync.   Now even if you powered them up, and they were magicly synchronized, they will probably not stay that way.  Slight differences in timebases, to different settings between units will cause them to vary.

Electrically you are putting the two outputs in parrellel.  This effectively halves the resistance of the load of each.   This is like taking your 4 ohm speaker and swapping it for a 2 ohm one.  Theoretically you have doubled the output needed, but the driver circuit can't deliver double the current, and clips the signal.  The output driver may not be protected and burn up, blow the transducer element, or just reduce the lifetime of the components.   

The case above assumes that perfect synchonization.  For the case of when one is transmitting and the other is receiving, you have a potential for the one transmitting to be doing so into a short circuit (through the other unit) or very low impedance, and the receiver being overloaded by too much signal.

Splicing the SI cable in itself is not an easy thing, which I have done.  There are twisted pairs of wires for each element with foil wrapping, wire for the temp and ground, with another layer of foil.  Grounds and foil (shield) are connected to the ground pin at the unit end connector.  Lots of wires and lots of different shields to mess up and allow noise to get in the system.   In my case it was just splicing a cut cable back together.  You would be doubling the amount of wires and shields, effectively doubling the chance for a connection problem.

[stuntman]

thanks FuzzyGrub

I agree that a physical splicing the cable would be a poor choice.  There are ways to listen in and interpret signals though.  High speed...they have got that...copper wire.

It sounds to me like they just didn't feel like doing it OR maybe they have and have just not released it yet.  I am certain allowing open map sources could be done even easier but that too doesn't fit into the business model.  The simpler solution was the A/B switch that can be sold for $80 (ugh).

Well it is good to see that this is starting to be introduced in the higher end units.  If it is working like I hear it is supposed to then it is a much better solution than having an external unit handle the SI.
Competition is good.

[FuzzyGrub]

thanks FuzzyGrub

I agree that a physical splicing the cable would be a poor choice.  There are ways to listen in and interpret signals though.  High speed...they have got that...copper wire.

A passive listening, intepretation, and simulating of the sonar signal back to the other unit would certainly cost more than you would want to pay and more than what a new unit would.

Having a copper wire and nothing to transmit or receive on, is not too high speed.  The existing comm channel is too slow and used by other devices, even if you wre only transmitting one row of pixel data at a time to the other unit. 

Personally, I'd like to see how it works with the new ethernet capable units.  Is there a video clip out showing this? 

[Humminbird_Greg]

FuzzyGrub pretty much nailed why this is impossible and I believe it mostly gets down to being able to synchronize the two units.  The pre 2011 798 units do not posses the capability to do so.  I don’t know how the Ethernet will work with this other than the fact that it is a very high-speed connection.

[stuntman]

impossible???     I will leave it at this    It is irrelevant anyways, I am not in school anymore and don't have time to start Digital Signal Processing and electronics and sonar projects.  For fun my guess on the retail for that item is $200-$250.....just a fun guess.

The video out is a great compromise on those older units (and new?) and the new networking sounds great and I am looking forward to reading more about it.

When I called humminbird about the new units I was told that the sonar sharing should work and may work between two identical units.  I  guess people will have to wait and see. 

My only remaining interest in this is for a friend who is buying a new boat and along with it new fish finders........I don't want to be left blind on the stern 
and shopping with/for a friend is at least as much fun as shopping for myself!

[Bob B]

If the new software and ethernet units are able to share sonar data, it would be at a different "level" than sharing the transducer directly.

The ethernet connection will be fast enough to send the data from one unit to the other AFTER it has been received and interpreted  by one of the units.

If you try to connect 2 units to the same transducer, you would be sending the output of one unit directly into the other unit as well as the transducer.  This is almost surely going to blow the output amplifiers in both units.

[jbwilli]

After reading the old threads on "what is the technical reason a single transducer signal cannot be sent to 2 displays" I would like to add a comment.

One reply stated this cannot be accomplished because when you turn on both units, both will send a 'ping' signal out and this cannot be eliminated. I disagree. On our transducer pin connectors, one pin sends out the 'ping' on 1 wore and another pin receives it on another wire. Seems to me, if you eliminate the out 'ping' pin from the unit you want to act only as a receiver, you could easily make that unit become a 'display' unit only and not have to go the expensive Ethernet cable route, which only the top of the line models have. I believe it is possible to use the very cheapest models out there and make 2 of them run off 1 transducer by making a special split cable.
Comments?

[Humminbird_Greg]

Welcome to the Side Imaging Forums jbwilli.

The transducer for your 385c has two wires in it.  One is for the transmit/receive connection and one is for the ground/return connection.  When the unit transmits to the transducer it does so with a high voltage pulse.  To do this it needs a complete electrical path which means a two wire connection (a transmit wire and a ground or return wire).  After it transmits the unit listens for returned sonar signals.  To receive these it needs two wires again for a complete electrical circuit (a receive and a ground wire).  The transmit and receive are electrically and physically the same wires and connections on the unit and the transducer itself.  Disconnecting the transmit wire/connection on a transducer also eliminates it as a receiver of the returned sonar as well.

[jbwilli]

Thanks for that information! I stand corrected, however....
Somewhere between the transducer and the lcd display there MUST be a decision device (diode/transistor/e-valve/etc) that removes the output 'ping' signal and sends ONLY the input return signal to the final display screen input. Otherwise, the final screen display would contain all the outgoing 'ping' signals and it would be unreadable! That is the electrical path point that needs only to be tapped and sent to the same path point location on a second unit to allow a person to connect 2 (or more) displays to a single transducer. Yes, it would have to be a modification built into the unit at manufacture. It would be simple, but something the mfg would not want to release..understandably.
The simple answer is to just use 2 units with different transducer frequencies!

[Humminbird_Greg]

jbwilli,
There is no decision device.  The unit transmits a ping and stops the transmitter.  It waits with the receiver on for the returned sonar signals.  It than starts the same cycle again.

Even if there were some sort of “decision device”, the second unit would still have to know when the first unit transmitted in order to compute the distance to any sonar targets (like the bottom of the lake).  Without knowing when the sonar was transmitted it cannot calculate how long it took the sonar signal to be returned.  An option to this would be to have the first unit send the time lapse information to the second unit but that would have to be over a dedicated communication line and not the transducer transmit/receive lines.  If this dedicated communications line was fast enough and had enough bandwidth you could call it an Ethernet connection…

[jbwilli]

Thanks again for the rapid reply!
I am getting a real education on the topic!
I guess the part I am still confused about is this - If there is only 1 wire that carries both the ping 'output' signal and the return signal 'input' (as you explained), then how is the 'ping' signal not going to show up on the lcd screen final output. Seems to me that the 'ping' out signal has to be removed or canceled out somewhere prior to the lcd final screen driver signal???. The final lcd screen display only needs an input, not an output...correct?

[Bob B]

It's all about the timing .......I used to fix radar and sonar years ago in the Navy.

For the unit to be able to interpret the signal it is receiving it must also know when the ping was transmitted.  The time difference between the ping and the received echo's tell the unit how far the echo'd return is away from the transducer.

Simply getting a return echo would be of no value at all by itself, since it is the time between the ping and the echo that is critical......Not to mention the fact that the output circuitry in a unit is designed for the impedance of a transducer.  The high voltage ping from one unit going into another units output would likely smoke the ouput circuitry in both units.

Try to connect two stereo amplifiers to the same set of speaker and then set back and watch them let the smoke out.......
you can't put the smoke back in once you let it out.

[Humminbird_Greg]

Bob, I see that you know of the importance of the “magic smoke” in electronics!