Random Pulses

Firmware A3030D01.vhd for ISL7 (A3030D) is debugged and running. It provides random lamp pulses for any interval and pulse length. For a view of the modulation noise generated by random pulses, see our previous post. The ISL7 stimulus length is the expected number of pulses. The interval length is the average pulse period in milliseconds. The pulse length as a fixed multiple 30.5 μs.

The A3030D divides each pulse interval into thirty-two sections. For regular pulses, the A3030D always initiates a pulse in the first section and only in the first section. For random pulses, the A3030D initiates a pulse in each section with probability 1/32. Thus the average period of pulse initiation is the same as the interval length. With pulses longer than one section length, the pulses overlap. Pulses that occur at the end of an interval can overlap pulses occurring at the beginning of the next interval.

With 2-s intervals and 100-ms pulses, we counted 100 pulses in 207 s. With 100-ms interval and 10-ms pulses, we used a recording of lamp modulation noise to count 559 pulses in 60 s. The figure below shows the distribution of the number of pulses we counted per one-second recording interval.


Figure: Distribution of Pulses per Second for Random 10-ms Pulses with 100-ms Interval. Average pulse rate is 9.3 pulses/s. For 32-s view of recording and spectrum, see here.
We appear to have a passable approximation to a Poisson distribution of pulse generation. When the pulse length is significant compared to the interval length, the average number of pulses drops because the pulses are overlapping. But we still get 9.3 pulses/s average for 10-ms pulses at 10 Hz.

NOTE: The ISL6 accepted the interval length in multiples of 30.5 μs rather than 1 ms, so the ISL Controller Tool 4.1 has a version selector that allows it to control both ISL6 and ISL7 devices.

Quiet Modulation

The figure below shows lamp modulation noise on the X input of an ISL6 (A3030C) for 20% brightness. This noise occurs even with no lamp attached and no X input leads.


Figure: Lamp Modulation Noise on X for A3030C with 20% Brightness, 50-ms Pulses at 10 Hz.
When we modulate the lamp, we apply a 1-MHz clock signal to the gate of Q1 so as to turn on and off the lamp current. During lamp modulation, the A3030C battery current increases by around 5 mA. We suspect that this 5 mA causes the 3V0 output of our TPS70930 to drop by 500 μV, which in turn causes X to step up by 2 mV.

In the A3030C, Q1 is the NDS355AN, with 200 pF gate capacitance. We load our A3030D with NTR4003N for Q1. The figure below shows A3030D modulation noise for random 10-ms pulses.


Figure: Lamp Modulation Noise on X for A3030D with 80% Brightness, 10-ms Random Pulses at 10 Hz. Vertical range is 140 μV.
The new transistor reduces modulation noise to 50 μVpp. The NTR4003N has resistance 1.5 Ω, which is much larger than the 0.2Ω of the NDS355AN, but still insignificant compared to our 60-Ω lamp lead resistance.

ISL7 Head Fixture

The photograph below shows our first ISL Head Fixture (A3024HFD-B). Version D of the head fixture has the guide cannula cemented parallel to the optical fiber.


Figure: ISL7 Head Fixture with Blue LED (A3024HFD-B).

This guide cannula may be used to hold the head fixture in place during implantation. Once the head fixture is covered by cement, the silica guide cannula tube may be crushed with scissors, or a thin pair of cutters, just below the thread. We now have a more compact head fixture, with no guide cannula thread protruding from it, and the hole in the animal's skull must accommodate only the optical fiber. With our previous head fixtures, the skull hole had to accommodate both the fiber and the guide cannula.

In the above example we have roughly 3 mm space between the circuit board and the guide cannula thread. We believe this is sufficient to allow cement over the circuit board and still have space enough to cut the guide tube. But we await comments from ION before deciding the final geometry.

UPDATE [24-JUN-15]: ION tells us they would rather have the older A3024HFC-B.

Command Reception in Water

We put a self-propelled ball in a latex glove and attach ISL A3030C No6.1 to the outside. We place the combination in a tub of water and let the ball propel the ISL around the tub, immersed in water most of the time. We place the tub in an FE2A faraday enclosure.



Figure: ISL A3030C No6.1 Moving in Water.

We have only one BNC feedthrough in this enclosure. We use this feedthrough for the data reception antenna, which we connect to one input of an Octal Data Receiver. We must open a hole in the enclosure wall to allow the command antenna cable to enter. This compromises the isolation of our enclosure. We add two more antennas by raising the lid a little, but this does not help much. The result is 92% reception of data messages transmitted by the ISL.

We use the ISL Controller Tool to request acknowledgments from the ISL in response to commands transmitted by our 915-MHz Command Transmitter (A3029B). Each acknowledgement is a single 915-MHz message transmitted by the ISL and received with 92% probability by the Data Receiver. We request 116 acknowledgements and receive 100. These observations suggest the reception of 915-MHz commands by the ISL's crystal radio while swimming in water is roughly 94%.

ISL Stage 6 Delivery

We ship to ION the devices shown below. There are five Implantable Sensors with Lamp (A3030C), five Head Fixtures (A3024HFC-B), a 915-MHz Command Transmitter (A3029B), a Loop Antenna (A3015C) for command transmission, and various cables and connectors.


Figure: Batch ISL6 Deliverables. The five encapsulated A3030Cs are on the center right. For a close-up see here. The ISLs are encapsulated in black epoxy and clear silicone. On the left we have antenna cable, LWDAQ cable, and boost power supply for the command transmitter, which is itself the square metal box. We have three BNC feedthroughs to enhance an existing faraday enclosure. Five head fixtures are arranged in petri dishes below.

Each ISL provides a 50-mm 915-MHz command antenna and a separate 50-mm 915-MHz data and acknowledgement antenna. Command reception by a moving ISL in a faraday enclosure at 915 MHz is >95%, compared to less than 10% with our previous 146-MHz command frequency. We will perform more exhaustive tests of reception at 915 MHz in the near future. The A3030C can respond to commands with an acknowledgement on the 915 MHz auxiliary data channel, without disturbing or requiring simultaneous data transmission. Version 3 of the ISL Controller Tool allows us to detect acknowledgements collected by the Recorder Instrument.


Figure: ISL Controller Tool, Version 3. To receive acknowledgements from ISLs, the Recorder Instrument must be acquiring messages, either with the Loop command in the instrument itself, or the Record command in the Neuroarchvier.

The lamp leads are orange (L+) and purple (L−). They are held together with silicone so as to reduce the magnetic field they generate when the lamp current turns on. The EEG leads are red (X+) and blue (X−). These are held together with silicone so as to reduce the magnetic field they enclose when the lamp current turns on. The input impedance of the A3030C amplifier is only 100 kΩ, compared to 10 MΩ in previous devices. These precautions together reduce the amplitude of the spikes induced by lamp current.


Figure: Lamp Current and Voltage Noise on X for A3030C with Lamp Pins Above Water, White LED Attached, and Lamp and EEG Leads Running Close Together. Pulses of 100 ms every 200 ms.

With the A3030B, the above arrangement of the ISL in water produced voltage spikes much larger than the 20-mV range of the EEG amplifier. The noise spikes are now of order 200 μV. We appear to have reduced the induced noise amplitude by a factor of one hundred. We see a square wave much larger than 20 mV if we immerse the lamp power pins in the same water as the EEG pick-up leads. During implantation, we must make sure that there is electrical isolation between the lamp pins and the rat's body. The pins should be entirely covered with insulating cement.

Outside the light pulses, there should be no noise induced upon the EEG input because the A3030C turns off its boost regulator between pulses. During the light pulses, after the initial spike when the lamp turns on, we see the same 8 μV amplifier-generated noise we see outside the pulses. Because of the way it manages its power consumption when idle, the boost regulator generates less noise when it is delivering a continuous current than it does when it is delivering no current at all. Thus we see noise spikes only at the start and finish of each light pulse.

One feature does remain in the EEG recording during and after pulses: a relaxation in response to the step-component of the lamp current noise, rather than its pulse component. We cover the LED so its light does not land on the water near our EEG electrodes, and the step-component of the lamp current noise remains unchanged. It is not a photovoltaic effect. When the lamp power turns on, the lamp current leads now carry +5V and 0V instead of both carrying 0V. Despite the insulation around the lamp leads, this change affects the potential of the ISL body, and so induces a step change in the EEG amplifier input of order 50 μV, which then relaxes in response to the amplifier's high-pass filters. With shorter pulses, such as those shown on the same scale below, the steps cannot be seen.


Figure: Lamp Current and Voltage Noise on X for A3030C with Lamp Pins Above Water, White LED Attached, and Lamp and EEG Leads Running Close Together. Pulses of 10 ms every 100 ms.

The A3024HFC-B head fixtures are all equipped with blue LEDs. We measure power at the tip of the fiber to be 13-16 mW at 40-mA forward current. The average is 14.5 mW. The forward voltage drop of the LEDs is 2.8 V, we have 5-V lamp power, and the lamp leads have resistance 60 Ω. When implanted, the current will be 36 mA, so we expect power at the tip to be 13 mW. This power is sufficient to provoke optogenetic response with 5-ms flashes at 10 Hz.

When the lamp is on, the A3030C draws roughly 60 mA from its 190 mA-hr battery. Stimulating with 5-ms flashes at 10 Hz, the battery can run for 76 hours. With the internal state machines running, but no data transmission or lamp stimulation, the A3030C consumes only 60 μA. Allowing for one hour of stimulation per day, it should be possible to run a 60-day experiment with the A3030C.

The goals of ISL6 were to provide reliable command reception inside our 915-MHz enclosures, and to reduce the lamp-current noise induced in the EEG input. We look forward to seeing how well the devices perform when implanted.

915 MHz Command Reception

We test the reception of ISL commands with our new 915 MHz Command Frequency. Our Command Transmitter (A3029B) is equipped with a 915.000 MHz 915 MHz Oscillator. This auxiliary circuit can stand alone, as shown below, or be soldered onto the A3030B circuit board as its source of 915 MHz.


Figure: The 915 MHz Crystal Oscillator (A3029XO-915). The circuit consists of a 457.5 MHz crystal oscillator, a frequency doubler, and an amplifier.

The Implantable Sensor with Lamp (A3030C) is equipped with a 50-mm data transmit antenna and a 50-mm command reception antenna. Both antennas are designed for 915 MHz when implanted in an animal, but as with SCT antennas, they work well enough in air as well. Now that we have the same frequency in use for data and command communication, the command antenna picks up the outgoing 915 MHz power from the transmit antenna, resulting in self-reception by the 915-MHz crystal radio, as shown below.


Figure: Self-Reception at 915 MHz. The 50-mm data and command antennas are parallel and separated by 2 mm. Top trace is the crystal radio output recorded with ×1 probe, 20 mV/div. Bottom trace is the 5-MHz data transmission clock recorded with ×10 probe, 1 V/div. Time is 2 μs/div. The data transmission lasts for 7.7 μs.

A self-reception pulse like this during command reception is likely to corrupt the incoming command. So the A3030C disables its data transmission during command reception and processing. Because commands begin with a minimum 5-ms pulse of 915-MHz power, the self-reception cannot be confused for command initiation, and once initiation has taken place, the self-reception is suppressed until the command is complete. Given that the 915-MHz command transmission will in any case overwhelm the much weaker data transmission, the suppression of data transmission during command transmission does not cause any additional loss of data.

We place a Loop Antenna (A3015C) at the center of a Faraday Enclosure (FE2B). We connect a white LED to our A3030C. We transmit a command to flash the LED once, and repeat this command at around 20 Hz. We move the A3030C around in the enclosure for 60 seconds and observe a total of 2 s of loss broken up into around six different intervals. We appear to have 97% reception of commands within the enclosure.

We work for several hours with the circuit, stimulating it, turning on and off the data transmission, and measuring the noise induced in the EEG amplifier by the lamp current pulses. We find reception within the faraday enclosure to be reliable. With the enclosure lid off, we can receive commands within roughly 100 cm of the command transmit antenna.

We have taken the well-establisehd SCT antennas and faraday enclosure and used them in reverse for ISL command reception. The A3015C loop antenna now transmits 1 W of 915-MHz power, and the 50-mm ISL command antenna receives this power. Meanwhile, the 50-mm ISL data antenna transmits 200 μW of 915-MHz power, and another A3015C loop antenna receives this power.

In the long run, we will be able to combine the two ISL antennas into one, and combine the two A3015C antennas into one, with the help of radio-frequency switches. The ISL can transmit "command received" messages through its data antenna, which means it will eventually be possible for us to know when a command has not been received. Just as we have multiple data reception antennas in the SCT system, we could have multiple command transmission antennas in the ISL system. If a command is not received from one command antenna, we can re-broadcast the same command with another command antenna.

Because the SCT 915-MHz communication works well for implanted SCTs, and because we can provide multiple command transmit antennas in the future, we believe we can rely upon 915-MHz command reception in implanted ISLs.

ISL Operating Range at 146 MHz

We compare reception of repeated commands by the A3024A and the A3030A. We fasten an example of each to a piece of cardboard.



Figure: Two ISLs Strapped to Cardboard for Command Reception Test.

The A3024A simply turns on its lamp when it detects radio-frequency (RF) power above a threshold. We set the threshold as low as we can. The A3030A turns the incoming RF signal into a sequence of byte instructions with a checksum at the end. When we send repeated commands to flash the A3030A lamp, the A3024A lamp illuminates while the command is being transmitted, and the A3030A lamp illuminates afterwards, provided the command has been received without corruption.

We use our 146 MHz Command Transmitter (A3029A) with boost power to produce the RF commands. We measure maximum range of reception, for a favorable orientation, and maximum range for reception in 95% of orientations. We also test reception with the two devices in water, as shown below.



Figure: Two ISLs In Water for Command Reception Test.

We measure maximum range for reception in water. Now we add 5 g of salt to the 500 ml of water and stir until dissolved. We measure maximum range again. We present our observations below.

Test	A3030A No1.2	A3024A No7
Robust Reception Range on Cardboard	70 cm	100 cm
Maximum Range on Cardboard	250 cm	600 cm
Maximum Range Held By Lamp and Dangling	300 cm	900 cm
Maximum Range in Water	70 cm	450 cm
Maximum Range in Saltwater	70 cm	450 cm

Table: Comparison of Reception for Encapsulated A3024A and Encapsulated A3030A.

It occurs to us that encapsulation or an exhausted battery might affect No1.2's reception, so we try an un-encapsulated A3030B with a fully-charged battery and measure maximum range in air. We get 250 cm, the same as for No1.2. We walk around with the lamp held in our fingers, and the device dangling below, and repeat our measurement of maximum range.

The A3030A must monitor and interpret the exact timing and content of the digital RF signal it receives on its two-turn loop antenna. Compared to the A3024A, which simply turns on its light in response to any received RF power, the A3030A has substantially shorter operating range (70% far in air for robust reception). In air, the A3030A's maximum operating range is one third that of the A3024A, but in water it is only one sixth that of the A3024A. In 1% saltwater, operating range is unchanged.

We place the A3030A in a jar of fresh water. It rests on the floor. We move it around 50 cm from the antenna and obtain robust reception. But there are orientations that are hard for us to obtain with the transmitter loose in the water, so we suspect that robust reception in all orientations would be closer to 30 cm. At ION, we observe robust operating range for an implanted device is only 20 cm.