Links and reading
Animated project intro
Simple EEG instructions
Main mailing list
The text and graphics of this site are released under a Creative Commons license unless otherwise stated.
Joe Street's active electrodes
Joe's active electrodes have the following main differences compared to Jarek's or Pedro's:
Some people may prefer this design, whilst others may be happier making the extra power wire connections to their modEEG box. It is up to you.
As at Dec-2004, only Joe has built and tested these, but he has been using them for nearly a year. Most probably other project members will try them in due course. If you are weighing up the various active electrode designs created by OpenEEG members, it might be good to check the mailing list archives for news, or to ask on the list if you need more info.
Note: Joe feels that his Eagle files need checking over before submission to Olimex or some other company for preparing boards, to be sure that they are fully correct. We will update this page when that is done. You could always etch your own boards, though, if you are comfortable with that.
Also, if you are the first to order boards from Olimex, please let us know the order reference code (via the mailing list) so that other people can save the setup costs on repeat orders.
Here are the current Eagle files for the boards and a bill-of-materials (in Excel format):
Please bear in mind the note above -- that the current version (Dec-2004) needs checking over before sending to Olimex.
Click for larger versions.
by Joe Street
I decided to try to make a battery powered version of Jarek's active electrode. I thought about the problem and decided a pin grid on a small board which could mount to the second board which had the amp and power supply was the best solution to my needs. I want these electrodes to be a plug in replacement for passive electrodes and give greater ease of use without making any modifications to modeeg. The idea here is to poke the four gold plated pins on the electrode board through an elastic headband to mate with four female receptacles on the amp board. This allows the electrode to be securely mounted to the headband and yet the electrode could easily be removed if I need to clean it or make a modification to the design. Here I used header strips for the pin grid but they don't need to be so long unless you have very thick hair! I am looking for header strips with shorter pins.
The completed electrodes are 16 mm wide and 32 mm long. The one on the left has a trial version of a cover I dreamed up and I left the other open for the picture. There is a little surface mount switch to save the batteries when not in use. I etched these boards myself for prototyping.
In the final shot the pins are not through the elastic so the electrodes can slide along the elastic headband. One can simply place this headband on the head, wiggle the electrodes to get firm contact and be ready for neurofeedback. It looks like the paste is history! The DRL cable still has a passive silver electrode but paste is not needed on this one when using active electrodes. I put the DRL to my wrist. Currently there are problems in running these electrodes in referential mode. I am still working on this but as they are here they work really well in bipolar pairs.
by Joe Street
1. Cut the header strip into sections of four. Four pieces are needed for each electrode to be constructed.
2. Install the header strip pieces on the electrode board and solder each pin from the other side.
3. Insert the gold plated male pins on the side which you just soldered the header strips but do not solder yet. You should now have an assembly with the four gold pins sticking up from the opposite side where the array of electrode pins is as can be seen in the background in Photo 1. Set the assemblies aside. The pins will be soldered after the electrode is plugged into the amplifier board to ensure the best pin alignment.
4. Begin with the amplifier boards by soldering the battery retainers in place so that the openings for battery insertion both lie on the same side. Which side you choose is not important but I made all mine consistent. Before you solder the retainer in position melt a little solder onto the pad in the middle of the battery area on each side so that it forms a slightly convex surface to contact the negative battery terminal. Position the retainer so the rectangular pads at each end sit symmetrically on the solder pads provided for this purpose and clamp it securely so that it doesn't move while soldering. Do both top and bottom battery retainers on each amplifier board.
5. Identify the bottom side of the amplifier board. This is the side which has the surface mount pads which match the terminal spacing on the amplifier. Insert the four large female pin receptacles from the bottom side and ensure they are fully seated. Now plug the electrode board into the amp board and check that all the pins are still fully inserted in the circuit boards. Solder the pins and pin receptacles in place on each of the boards. Carefully remove the electrode board being careful to keep the two boards parallel while removing to prevent bending the pins. Set the finished electrodes aside.
6. The next step is to solder the surface mount slide switch in place on the top side of the amp board. The orientation of the switch is not important but again I did it consistently on all my boards. Use a soldering iron with a very fine point. The technique I find works well is to position the part on the board and then clean the iron tip and melt a small bead of solder on the iron tip. Hold the part in position and touch the iron tip to one corner leg of the switch just until the solder flows from the iron to the leg and pad. Remove the iron and continue to hold the part steady for a couple of seconds until the solder is solidified. I do the opposite corner next and ensure the alignment is good as well at this time. The remaining pads can then be completed in the same manner but the part will hold itself in place. Photo 2 shows the detail on the amplifier top side.
7. Soldering the amp is similar to soldering the switch but due to the smaller size it requires a delicate touch. Observe the correct orientation and be careful not to hold the iron on a leg of the part for more than a few seconds to avoid damaging the little amplifier with too much heat. After soldering a few of the switches you should have a bit of a feel for the process. When you have the amp board bottom side up on the bench in front of you with the long axis going from left to right and the battery clip on the right, the writing on the amp IC should read from left to right ensuring that pin 1 is to your bottom left.
8. The last step is to add cables to your active electrode. With the amp board oriented as in step 7 the signal output is found at the pad at the lower left corner of the board and this should be connected to the signal carrying conductor for your channel input while the two pads at the upper left connect to the shield. It should be noted that the shield of your cable must be connected to VGND at the input to the EEG unit for these electrodes to function properly. I use RG-174 small diameter coaxial cable for my cables. Trim back the shield leaving a couple of mm exposed to make the ground connection at the two pads on the upper left. Take a small section of stranded wire that fits these pads and wrap a single turn of it around the exposed section of shield and poke both ends through the two pads and twist together on the top side if the board. Now carefully solder them to the pads on the top side. On the bottom side flow a little solder onto the stranded wire where it loops around the shield and let it flow into the shield conductors a little to anchor the cable firmly to the amp board. Solder the center conductor at the pad on the lower left. Put the appropriate connector on the end of the cable to mate with what exists on your EEG unit and you are done. See Photo 3 for details on the amp and cable mounting.
9. A word of caution. This active electrode was conceived as a minimal parts design and as such there are a couple of caveats to be aware of. There is no protection against reverse battery insertion. To allow the maximum headroom for the +/- 3 volt battery supplies , protection diodes were not included. Pay strict attention to battery orientation as marked on the battery holders to avoid blowing the amplifiers. Similarly ESD protection was not incorporated in the interest of simplicity and small size. On the other hand if an amplifier is ever damaged it is a simple matter to replace it. Having said this I have been using four electrodes now for almost a year with no need to do any repairs.
10. The spacing of the pins was selected to be used with 25 mm wide elastic strap. The electrode pins can be driven through the elastic to lock the electrode in position on the strap. Alternatively I have found that it works well by just placing the elastic band between the pins so that the electrode can be easily and quickly repositioned along the length of the strap. They still hold their position well this way. The decision to route the wire off the side of the amp board was made in order to allow the cable to be attached to a plastic slide which is made to fit 25 mm wide webbing to act as a strain relief. I used colored cable ties for this purpose and also fitted another cable tie of the same color to the plug end of the cable to facilitate channel identification. The switch is slid toward the batteries to turn on. Do not forget to turn off when not in use and the batteries will last a very long time.
11. In use the electrodes are very simple. Make sure the pin grid is in good contact with the skin with the entire grid. The skin should be clean and not too oily. The reference electrode can be a passive one on the wrist, and should be connected to VGND. I use silver and have not had a problem with galvanic charge due to the difference in potential between the gold pins and the silver wrist electrode. Allow a few moments for the electrodes to stabilize. I have experienced stability problems when trying to use DRL. It can be made to work with careful adjustment but the signal quality is no better than using VGND as a reference, which gives excellent results.