I am pretty new to electronics but so I've been doing a lot of reading and whatnot. I have a question I can't seem to find any information on. I am modding a handheld device (the device in question isn't important) and this involves soldering to several spots on the motherboard which are DAT0, DAT1, Dat2, Dat3, GND, CLK & SMD (these grant access to the internal storage nand chip).

Each of these connections are their own wire (7 wires total) that all lead to an SD Card reader PCB. The problem is, I want to be able to turn on/off all 7 connections using just one switch, because when these wires are soldered to the SD Card reader PCB, the device boots into a special recovery mode, but I only want the device to boot into that recovery mode when a single switch is flipped on. I need a way to break the connection on each wire in between the device's motherboard and the SD Card Reader PCB (and be able to connect them up again using a single switch). I think that would be so cool!

I've considered using a series of small surface mounted solid state relays in parallel, wiring each of the 7 connections through the collector and emitter side of each relay. However, I'm seeing that solid state relays are typically used to close a circuit with a higher voltage, like 60V or something even higher. Are solid state relays ever used for data connections on the collector and emitter side? Would using a solid state relay somehow ruin the flow of data from the device's motherboard to the SD Card reader's PCB? I'm not asking anyone to come up with a schematic or anything, but if there's some other kind of small smd component that would suit this task better, please let me know!

Also, please keep in mind I need these components to be small enough to fit in an area about 4mm high! I would totally use mechanical relays, but those are far too big!

Sorry if this post is confusing, I was too tired to sketch out something, but if that would help I'd be glad to do it as well as answer any other questions you may have. Thank you for taking the time to read!

  • 1
    \$\begingroup\$ How you could think that a schematic wouldn't be required is beyond me. \$\endgroup\$ – Andy aka Jul 2 at 8:47
  • 1
    \$\begingroup\$ @Andyaka I was actually pretty excited to see this comment until I realized you decided to be rude with your response. I get that there are probably a lot of questions you see that seem stupid or unnecessary, but you'd think as someone as experienced as yourself would recognize another person getting started trying to learn for themselves, and maybe give them a positive nudge in the right direction instead of whatever this was. I hope you decide to leave better comments for people in the future. \$\endgroup\$ – Drew7 Jul 2 at 9:07
  • \$\begingroup\$ @drew7 it is a perfectly valid assumption to at least provide some sort schematics or sketch of what is it that you are trying to do, and it is a perfectly valid thing to request this kind of diagram, as it always helps. When people ask questions, they always leave out important information for some reason, maybe because they think it is not necessary, or think it is not needed, or whatever the reason (too tired). And we simply don't get why people omit information, because it helps us to understand the question better, because sometimes the answer needed is not what the quesition is about. \$\endgroup\$ – Justme Jul 2 at 10:51
  • 1
    \$\begingroup\$ @Justme I do understand now that using schematics should be a thing I do regardless of how simple I think a circuit is, but that wasn't the reason for the comments above. The issue was the way some people prefer to drag others down because they could be doing something better, instead of giving respectful and insightful advice- it's even in the rules for the website. To me, it didn't seem necessary to include a schematic because I believed my question was simpler than it turned out to be. Anyway, I do appreciate your more respectful comment and answer, unlike some people around here. Cheers! \$\endgroup\$ – Drew7 Jul 2 at 11:11
  • 1
    \$\begingroup\$ @Ilya Yeah I think this is what I will try first. It's worth a shot. Thank you! \$\endgroup\$ – Drew7 Jul 2 at 12:30

SD and eMMC signals are high speed, low voltage signals and messing with them could ruin the signal integrity. It's not possible to switch them with a mechanical or solid state relay.

The device you might be looking for is an analog mux or switch.

| improve this answer | |
  • \$\begingroup\$ Ahh okay, thank you for the suggestion! \$\endgroup\$ – Drew7 Jul 2 at 9:18

@Justme is correct that the signals MAY be affected by adding external connections (it's part of the risk of our not having all the available information to work with), but there is a reasonable chance that a 'high impedance when off' connection would be acceptable. Balance between circuits MAY be important.

Solid state switches may be usable but without information from you re signal voltages, signal speed, impedances and which if any connections are bidirectional it (again) gets hard to provide a good answer.

A solution which has a reasonable chance of success is to use "magneto-mechanical" reed relays, or reed switches operated either by a user wielded permanent magnet or a user wound activation coil.The smallest reed relays that I could found are in the 5+mm height range, but reed switches are available from around 2mm high.

As an example of one of many possibly suitable reed switches, These reed switches are 2 mm high, leaving space for a wound coil within your 4mm height limit.
$US0.50 each in 1's in stock at Digikey.
Also similar and
Many more reed switches

If coil-activated they require in the 10-20 ampere-turn range of activating current - so eg 200 mA and 100 turns or 20 mA and 1000 turns or ... .

A potential issue with any mechanical switch is contact resistance and potential need for "wetting current". If contacts carry no DC current they may exhibit a high contact resistance and may need a minimum voltage to break down contact surface oxidation. Sealed reed relays are less prone to this than open contacts but it is an issue that one needs to be aware of.

It may be possible to create an actuator that uses a solenoid or linear or micromotor to move a permanent magnet.


4mm dia pager vibrator motors here

Some of these

enter image description here

| improve this answer | |
  • \$\begingroup\$ This is super helpful, thank you! I'd give a check mark to both of you if it would let me haha. \$\endgroup\$ – Drew7 Jul 2 at 11:38
  • \$\begingroup\$ As this is about EMMC and SD storage, the signal speeds, voltages and bidirectionality are known. At least 20 MHz, up to 200 MHz, 1.8V or 3.3V, clock is unidirectional, command and data pins are bidirectional. \$\endgroup\$ – Justme Jul 2 at 11:39
  • \$\begingroup\$ @Justme Thanks. If the OP knew this he should post it or a link to a reference. I was aware that I could search for such info - and I'm pleased that you have supplied it - the OP realising he should do so is a valuable part of the learning experience. I appreciate the tensions on both sides of the recent comment exchange. Your input is fine. I wish others would be a little more tolerant of each other :-). | I think reed relays with minimal lead length on the card side and best attempts at balance on the switched side "may even work" . Or not :-). \$\endgroup\$ – Russell McMahon Jul 2 at 11:49
  • \$\begingroup\$ @RussellMcMahon Thanks for the kind words! I'll definitely focus on ignoring those other comments from now on. \$\endgroup\$ – Drew7 Jul 2 at 12:11
  • \$\begingroup\$ @Drew7 ... while noting any good advice mixed in with them :-) :-) \$\endgroup\$ – Russell McMahon Jul 2 at 12:13

Not the answer you're looking for? Browse other questions tagged or ask your own question.