Ultrasonic Module - Teardown part 1

Projects / Project Swiftlet  Original post date: 05/22/2015

I guess I'll order some modules to play around. The prices are lower than what I would pay for the Ultrasonic transducers. Let's see how much reverse engineering I can do without having ones of these in my hand without waiting for 3 weeks or so.

It is nice of DX to link a datasheet (might trigger security warning) and sample code. ElecFreaks has english user guide, demo and blog article.

• Model: HC-SR04
• Working voltage : 5V(DC)
• Static current: Less than 2mA.
• Output signal: Electric frequency signal, high level 5V, low level 0V.
• Sensor angle: Not more than 15 degrees.
• Detection distance: 2cm~450cm.
• High precision: Up to 3mm
• Mode of connection: VCC / trig(T) / echo(R) / GND
• Shipping Weight: 15g

They have a reasonable range, not bad on the power consumption (15mA during transmit) and should be easy to use. On the other hand, I don't want to deal with 5V if I can help it.

I can turn off the power between samples to save the 2mA quiescent current.

8 * 1/40000 * 15mA x 10 per sec x 2 averaged over 1 second is below 100uA.

There is a SGM324 made by SG-Micro on the left hand side behind the receiver. I haven't even heard of the company until today. These are not your average "324", but are CMOS rail to rail I/O quad opamp. Gain Bandwidth Product is 1MHz, so we should expect at most a gain of 25 at 40kHz from each of the individual opamp. Depending on the gain requirement, I might use a LMV324 as they are cheap and easy to source or use something with a higher GWP.

Ultrasonic module interface timing

The middle chip has no label. There is a 4MHz crystal Y1 connected to it. It could be a microcontroller that generates the 40kHz signal for the transmitter and filters out the received signal.

I found this on Banggood. Probably a picture of a old version as the design is different and the chip number is still there. There are parts that got removed.

Unfortunate I don't have the usual zoom-enhance-pan trick on CSI. This is as far as I can clean up the picture.

If I had to build a transmitter driven from a logic level, I would have used a MAX3232 to drive it at +/-10V p-p. It has a charge pump for the +/- 10V, level translator and drivers that are designed to drive big capacitive load e.g. the 2200pF in a transducer. The 5 capacitors near the 16 pin chip is what you would typically use for the charge pump and decoupling. Pin 7 attached to the transmitter is a transmit pin on the MAX3232. They could be using a version or a clone with internal charge pump and shutdown pin for the later revs.

Driving a transducer with 2200pF at 40kHz is just north of 10mA, add in the extra current for the rest of the board, that's just about the 15mA transmit current quoted.

-----------------------------------------------------------------------------------

Emil has done reverse engineering here. I am right about the MAX2232! They had replace the awful LM324 with the SGM324 and probably use a different microcontroller in the recent versions.

Judging from the ebay/Chinese sites, the most likely part# that is on the module is: TCT40-16R/T-2 from Sunway / Shenghu (might trigger security warning)

1.5ms ringing is about 51cm.