Jump to content
Terry

LMH6574

Recommended Posts

Im finding it hard to get a good quality output from this chip, has anyone here used it ?

Terry

Share this post


Link to post
Share on other sites

I've never used it, but the basic specs look good. If you are having problems, my first thoughts are that maybe you are trying to operate it on a single voltage supply or the gain is not set right. Can you share your schematic and some photos of the construction?

Share this post


Link to post
Share on other sites

Yep, I think you hit the nail on the head, I was using a single voltage supply, I was trying to check this from the data sheet but could not find what I wanted. It was only when I found a copy of the evaluation board that I realised. Can you recomend any other 4:1 video mux that works from a single supply ?

Terry

Share this post


Link to post
Share on other sites

I don't think any are compatible with a single supply since video is AC coupled. However, there are ways to trick things. In my MAXIM based Video Switch, I DC biased the video inputs so that the AC coupled video was at 1/2 the single supply voltage. But, the usual solution is a bipolar supply.

Share this post


Link to post
Share on other sites

I use the 4311, have a little stock of those that I got for free... requesting 2 samples each month along with some other things works well :P

I've powered it with a single-supply, and biased the input. Maybe do a little search, I know I've given the details somewhere on here.

Share this post


Link to post
Share on other sites

Uh. I was mixed up, thought you were talking about the buffer rather than the switch. The schematic that stands in the link you brought back didn't care about bias. That was working fine with the gear I was using at the time, but it's not the case with all cameras. It would be better to add a 680 Ohm res to the power supply, and a 82 Ohm one to ground at each input you intend to use.

Share this post


Link to post
Share on other sites

It seems to me that the schematic in the link you posted will require that the video be DC biased well above ground. Not all video sources will comply with that since most are AC coupled. So, I'm not sure it will totally cure your problem unless you make some design changes. It sounds like Kilrah has made some suggestions in that regard (but I think it will demand something more to be fully robust).

Share this post


Link to post
Share on other sites

The suggestion is what I use in my video buffer. It has so far worked with all the sources I've used it with... but indeed the switcher caused problems when I changed the stuff I had when I made it. I didn't really know about the video signal at the time :unsure:

Edited by Kilrah

Share this post


Link to post
Share on other sites

Hmm, I was hoping this would be the easy bit but it seems not. I will have a go at getting it to work but if its taking too long I will use relays for now. Just until the rest of the project is working then maybe come back to it. Is the DC bias the main problem ?

Terry

Share this post


Link to post
Share on other sites

Ive just been playing with the 6574 and things are looking up. I have used 2 resistors to make a voltage divider to supply it and its starting to give useable results. I will go ahead and make the propper board for it now.

Do you guys have any advise on layout ? Do I need to use a double sided board and if so are the grounding points critical ?

Terry

Share this post


Link to post
Share on other sites
Is the DC bias the main problem ?

In a typical video environment, the video is AC coupled. So, much of the useful signal can swing below ground. That is why the video mux would be operated from a +/- supply (otherwise the signal will hit the gnd rail and corrupt the video).

However, some video cameras have omitted the output cap (not a good idea, but it saves costs) and they will have a dc level output set well above gnd. This can help your LMH6574's situation. You could check your cameras with a scope for this. If that is what you find, then a single supply solution will work.

Do you guys have any advise on layout ? Do I need to use a double sided board and if so are the grounding points critical ?

Just treat the pc layout like a RF design. Composite video requires several Mhz of bandwidth. To ensure full video fidelity, the layout must be done with care. Signal routing and grounds need the usual RF attention for best results. I have seen some marginal pcb designs that fully satisfied the user's video expectations (but not mine), so the need for a precise layout can be cheated a bit.

I will have a go at getting it to work but if its taking too long I will use relays for now.

Relays are definitely easy. However, you may find that the video switching will look a bit dirty for a moment after you change video sources. A fast solid state solution will eliminate this.

Edit: If your are designing this for personal hobby use, and only need a 2-CH video mux, then I highly recommend this project: http://www.rc-cam.com/rcswitch.htm. It uses a single supply (using a bit of circuit trickery) and offers glitchless video switching transitions. I no longer have the PCB files on my PC (archived them long ago), so you'll have to create your own layout. BTW, it would also be a perfect use for one of your Bit-Switch chips. :)

Edited by Mr.RC-Cam

Share this post


Link to post
Share on other sites

Yes the idea of using a Bit switch chip is in my mind for an on board video switch but this is for another project. I have been testing my new plane and it out runs my auto tracking patch at close range so I am working on a new idea. The auto tracking patch will be upgraded for long range use only and I intend to auto switch between 4 patches set at 90deg for close use. Early days but it should work well I think.

Terry

Share this post


Link to post
Share on other sites

If your diversity Rx design will use the Airwave Rx modules, then I think you will be in luck. Their outputs are DC coupled. Be sure to install the big video coupling cap on the output of the MUX (instead of on the four inputs).

It would also be a good idea to wait for the Maxim chip. It will probably offer much better single supply performance than the method you plan to use with the LMH6574. Maxim's samples are shipped very fast... at least the USA deliveries are quick.

Share this post


Link to post
Share on other sites

I got the MAX4311 chips but they are tiny ! Seems I may have got the wrong ones.

Anyway Im still going to try and get the LMH6574 to work if I can but Im unsure about the A0 and A1 inputs. They show them connected to ground through a 50 ohm resistor but this can not be right can it ?

Terry

Share this post


Link to post
Share on other sites

I didn't see that they were specifically using 50 ohm terminations on the mux address (Rt), but the low value is not unusual. This IC is designed for very fast switching. So, the engineering test app circuits would be setup for highest performance. In this case, a carefully devised 50 ohm driver allows for testing the part at the rated 8nS access speed. In your case, your switching is at a snails pace, so the terminator can be omitted. The usual TTL interfacing is fine.

Share this post


Link to post
Share on other sites

I have just noticed that the video from this chip lacks contrast when compared side by side with my other receiver. I had not noticed before as I used it on its own so probably just adjusted the screen without thinking.

Not being an expert on things video could someone look at the data sheet for me and surgest how I may add a pot to adjust the level or failing that what resistors values I should change to.

Thanks, Terry

Share this post


Link to post
Share on other sites

Contrast is related to dynamic range. Can you post a schematic of the circuit with component values? Maybe an innocent mistake was made.

Here are my personal tips. Follow these and most of the common problems should be avoided.

1. Vid-In should have 75 ohm terminator to Vid gnd.

2. Capacitively couple the Vid-In to the mux IC with 330uF.

3. Mux amp should be set for 2X gain.

4. Mux amp outputs should have 75 ohm series terminations.

5. Mux amp outputs may need 330uF series caps, but it depends on what you connect to them.

6. PCB layout should consider the signal paths as a 10MHz RF design.

Edit: Thinking back on what I sort of know about your implementation, some of these suggestions will not be compatible with your design. I probably should have seen a schematic first.

Edited by Mr.RC-Cam

Share this post


Link to post
Share on other sites

I copied the one from the evaluation data sheet but swapped the 50R resistors for 75R. The psu is +5v -5v.

Eval data sheet HERE

Main data sheet HERE

Thanks, Terry

Share this post


Link to post
Share on other sites

Oh I forgot to say I have no 330uF caps on the inputs but a 470uF on the output.

Terry

Share this post


Link to post
Share on other sites

If that is the case, then my list of suggestions is good to go.

Item #2 is probably involved with your problem. Perhaps you are DC coupled to the Airwave module? With its offset bias, and the 2X gain of the mux amp, you could easily hit a V+ supply rail with bright whites.

Your scope will also offer a clue to what is going on.

Share this post


Link to post
Share on other sites

I got simillar problems. Especially with power supply. I haven't worked out the solution yet. So maybe you could help.

I bought two chips

MAX454 for video switching to work with the 5v transmitter + 11v cameras

and

MAX457 for splitting to work with a 5v receiver

I've been planning to power both circuits off lipo batteries.

Chips' applications are quite simple except the power supply, which must be +/- 5V.

So...

say I want to put MAX454 in one circuit with the video transmitter which requires 5V (and a 11V camera, which works well on 3s lipo). I use the full 3s battery voltage, add a 2 resistor ladder so that the signal gnd will be at 5V and supply the transmitter from 10V battery. Right?

But what next? How should I connect this to the transmitter that needs +5V only. If I connect it in the same way (i.e. keeping signal ground at 5V), will it still work and will I the 2 resistor divider suffice? Then comes the lcamera, which needs the full voltage from my battery.

Is it at all possible to achieve what I want, or maybe shoud I instaed buy a 4s battery and try to: power camera from upper 3 cells (0-11V); power MAX from cells no. 1 and no. 2 (+/- 3.7V) and transmitter from the 2nd and 3rd cell (7.4 V reduced by a regulator)?

You see, I am quite lost with this

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×