Interfacing fish finder take 2


I read this post and try to bump it to keep it together but it is to old and was not solved.
Interfacing fish finder "aborted..."

I want to interface the same Lucky fishfinder as the older post on arduino (200 khz transducer / 45 Degree / max 100m). I openned the handheld device with the screen and realize that reverse engineering will not be possible. I have another idea. This sonar come with a lcd screen and the screen is driven by this chip: HT1621B datasheet
I am thinking to connect the arduino on the lcd driver and take the data I want out of it to the arduino.
This way I could use this fishfinder in my ROV.

My problem is I have no clue of where I can take the data out of the HT chip and if I will need to condition the signal for the arduino?

Can someone help me with this? As you can see on the picture, I have access to the whole thing.

Thank you

In case it is important here is a picture of the other side

It takes a lot of work to reverse engineer a direct LCD driver interface, but this is how to start the process.

Less creepy link to a datasheet for that part:

Do you think that the data are produce in the HT or maybe the HT only receive the numbers from another chip and just drive the LCD? In such case, my point would be to connect on the data in pin of the HT not on the data out to lcd pinssssss.

In the datasheet, pin 4 is named "data". If I hook up there, I might receive the data directly?

The HT is only a seg/com driver. They're sending it pixel data from "the brains of the operation" -- another MCU.

What are the other two ICs on that board?

The HT turns on and off LCD pixels that can be any shape or size and located anywhere on the LCD, similar to how 7-segment displays work. All of the intelligence is in the microcontroller.

The only way to get data from the LCD lines is to follow the reverse engineering procedure linked in reply #2.

The two other IC are:

AA32416A (small one on the top rigth) Datasheet
From what I can see, the metallic square that look like a potentiometer just beside is probably the "Ceramic discriminator" mention in the datasheet. Some sort of reference for the sonar signal I guess?
If this is the case, maybe we can have the "clean signal voltage" out of pin 9 of this IC. They call it the demodulated signal.

I will poke around with a DMM tonigth and try to figure out what is connected. From what I can now, pin 5, 6 and 14 are visibly connected.

The other IC is : N76003AT20 page 11 of this datasheet

One last thing, on the first picture, we can see 5 solder pad in the lower rigth part. They all seem to go to the N76 IC. If we are lucky it is for QA testing and data could be monitor there? Maybe I will see something on a serial monitor?

To Blackfin, thank you for the LCD link. It sure look hard to do. I hope I will find a different way but it's good to know there is an alternative outhere.

So the N76’ is an 8051-based micro and is the brains of the outfit, so to speak.

TBH, I don’t think the raw data you’re looking for is going to be broken out anywhere. The microcontroller will likely just drive the transducer circuit, process the data internally and clock pixel data to the display driver.

The 5-pos connector footprint may be for a serial port, used for factory testing or it could be, say, a programming port. You’d need to do connection checks between each pin of the connector and each pin of the micro and compare that against the controller datasheet to get a better idea of what it’s for.

TBH, I don't think the raw data you're looking for is going to be broken out anywhere. The microcontroller will likely just drive the transducer circuit, process the data internally and clock pixel data to the display driver.

Yes this is probably the case and if your rigth I'm back to the LCD link I guess.
I don't understand how they can build this device and sell it 40$ while buying a very basic low market sonar garmin lowrance etc is so much expensive?

Is this correct? From my basic point of view the building block for a sonar are :
DC to AC converter -- AC pulse (PWM?) - high impedance transducer -- AC echo --> AC to DC conversion --> DC voltage to depth value.
The AC pulse frequency must match the transducer khz (I guess?) In my case 200 khz.
Somewhere in the process, a timer is needed to control time between pulse and echo. Or the transducer frequency already integrate the time?
Also need demodulation to remove any AC noise out of the transducer signal and convert to DC. Like a high pass low pass audio filter + DC conversion?

Maybe I can do something with the output DC voltage on its own?

If you can reverse engineer the circuitry between the transducer and the microprocessor, you could add another micro to capture the return pulse, then do the timing and display yourself.

I took a different example of the same "brand" of fishfinder apart, and was daunted by a separate small circuit board that has at least 3 tuned circuits, and several different connections to the micro. That was too much of a challenge for me, but yours looks a lot simpler. You will almost certainly need an oscilloscope to trace the signal paths.

Please post a link to where you bought yours. I would be interested in taking a look at it.

I found this manual, which implies the picture on the display can be quite rich. It is not clear, though, if those are predefined on the display or are generated dynamically by the MCU out of pixels.

If those are pixels, interpreting the display is going to be quite a challenge.
I was also impressed with the claimed range. 100m under water !

I don't know much about LCD. I only know that the depth appears a the top rigth corner. The fish, weeds and stuff may be build out of numerous pixel?

About the 100m claim, I work on reservoir and was absolutely impress by the sonar. It indeed measure up to 100m. I used it to measure depth of 3-4 meter up to 98,7 m with it and confirm the depth with a water quality probe and 2 other Handheld sonar. If we hack this thing I guess a lot of user will be happy to use it.

The claim that less impress me was the fish finding part. It is named "lucky" and based on my survey, it is lucky indeed since I found fish all over the place in most measurement. As expected, it also return a lot of false triggering over trees (that are submerged in hydro-electric reservoir) but all sonar do that.

I bought this unit on a**zon. Just try " LUCKY-Handheld-Portable-Fishfinder-Transducer " and you should find it.

To open the unit, you first have to unglue the black part that surround the LCD screen. Then you have acces to the screws.

I think the fish icone light up all together. There is 5 predefined fishes per rows and weeds or rock at the base.
Is this dynamical ? I don't know?
To me it look like the good old car racing game were you have to avoid obstacle on three lanes. You know, the good 1980 handheld mini-game.

Looking at this:

I question its effectiveness. You can see why it's quite cheap.

This one:

looks much nicer though on Amazon it's close to twice the price.

Looking at this:

Lucky fish finder is scam tested - YouTube

I question its effectiveness. You can see why it's quite cheap.

This one:

looks much nicer though on Amazon it's close to twice the price.

Yes As I mention, the thing see fish all over the place but for me it do not matter since I am looking for depth.
In the first video, they report a whole lot of fish. We can also see on the corner that the depth is stable at about 1.5m. Not way off from what I can tell by the image.
The wifi version is nice. We can also suppose that all the sonar stuff are in the wifi ball not in the handheld meter. This migth be a good way to have access to the sonar circuitry alone? For twice the price you need to be sure it work.

I finded time to look into the circuit in more details. It is too big to be ruled in one post... I would like to share were I am now before I go further. your input will help understanding the next step.

1- The LCD driver is essentially driving the LCD. I think it also output a clock signal to the transducer (switching the transducer accumulating capacitor maybe?) at a certain frequency. Thus triggering timed pulse. There is 4 com and 31 segment. A big job to hack with the arduino trick.

2- I tried to connect to the microcontroller with a serial device on the soldering pad. They are connected to RX / TX of the microcontroller, V+ and gnd. But I was not able to see or trigger any data transfer event. I used realterm and a rs232 cable "spy". Rx or tx was not blinking or doing anything.

3- The AA32416 IC (on the right corner) is were I choose to invest time for now: I figure out all the IC pin connection and this is were I need your help to go further.

Of the 16 pins, only 7 are connected and strange enough, pin 9 (IC final output pin is not connected - or I failed to find the link yet). I was expecting a lot from this pin.... (: What this IC do with only 7 pins? Can someone figure this out with my finding below?

Pin 4 and 15 are respectively Vcc (regulated 3,6 volts) and ground.

Otherwise pin 5 and 6 are individually ground through capacitor (18 pF each).
Those are The IF amplifier input pin and IF amplifer bypass pin. (see picture of the typical schematic from the IC datasheet).

Pin 8 (filter amplifier input pin) is connected to V+ through a 10k resistor.

Pin 12 (RSSI output pin) is connected to ground through a parallel arrangment of a 18 pf capacitor and a 1M resistor and it is the only pin that is connected to the microcontroller (pin 1).

Pin 14 (noise detector output pin) is connected to ground through a 18 pF capacitor.

I measure 18 pF for all capacitors. Probably it is wrong because all the part are interconnected? I probably read parallel capacitance of the circuit and this is why it is constant?
Finally, can someone explain my the basic of coil inductor (the 2 metallic square coil on the board). I google about this. They have 5 pin. 3 on one side and 2 on the other. I think the 3 pins are like potentiometer that allow to adjust the coil inductance value. The other end is fix and will act like a transformer? In any case, both coil inductor are not clearly connected to an IC. There is some transistors in between. I try to follow the signal path but my misunderstanding of the 5 pin vs the transducer input make this job much harder. I also think that the inductor on the rigth is only for reference signal since I don't see any connection with the transducer while the other one is connected to the transducer.

thank you

AA32416.pdf (419 KB)

It sounds like you are on the right track.

The aluminum shielded coils are most likely tuned impedance transformers. They could be used for the echo receive input circuitry and/or to increase the transmit voltage.

Do you have an oscilloscope, to follow the signal path? That would be a great help. It should be easy to identify the transmit pulse, but much harder to follow the return signal.

Unfortunately I have no scope. This tool is expensive. I am not sure I want to go there.

Scopes are absolutely invaluable. Borrow one, or ask a friend with one for help.Even a very low end scope is fine for audio/ultrasonic work (e.g. 10 MHz bandwidth).