Another Electronic Det Can Example
 

It's finally time to hit the dyno to tune spark. Speaking with the dyno operator (MER in Cresson, TX -- big time Miata race shop), I asked about knock detection while tuning. His answer . . . "we'll just use the existing knock sensor on the car." Uh oh.

My car is a 1990. It didn't come with a knock sensor. We can add one, of course, but you don't have to search much around here before you realize that det cans and human hearing (evolved over millions of years to dynamically detect minor sound variations) is a much better way to go. So, time to build a det can.

Now the question is, which type. There seem to be three main variations:
1. "Sound-Powered Phone" type. These are purely mechanical devices using a crushed piece of copper tube, flexible (but not too soft) air hose, and hearing protectors. If you're not handy with a soldering iron, these are a good way to go.
2. Electronic Stand-Alone type. These are electronic and use an electret microphone. Usually constructed by purchasing an off-the-shelf personal listener and then modifying it by adding wire to remote-mount the microphone in a battery clip. You listen with an ordinary headset.
3. Electronic PC type. Similar to #2 with the variation that you use your tuning laptop's built-in sound card to amplify the microphone signal. Again, an ordinary headset is used for listening.

I decided on option #3. These are the advandages as I see them:
1. As I'm tuning MS, I'm going to have my laptop on hand anyway.
2. The sound card built-in to my laptop is (I think) more high fidelity than an inexpensive personal listener.
3. If needed, I can use my laptop to record and analyze the audio.
4. I can use robust, shielded manufactured cabling.
5. This will actually cost less and be less work to build.

So, here we go:

Parts (PNs and prices are from Radio Shack -- use the vendor of your choice):
1. 1/8" (3.5mm) Stereo Plug -- PN 274-1547 -- $4.19:
https://www.miataturbo.net/attachmen...ine=1321038673

2. Electret Microphone Element with Pigtail -- PN 270-0092 -- $3.99:
https://www.miataturbo.net/attachmen...ine=1321038673

3. Battery Clip (2 to a package) -- PN 270-0343 -- $4.29:
https://www.miataturbo.net/attachmen...ine=1321038673

Total . . . $12.37.

First, solder the microphone to the stereo plug. The white microphone wire goes to the plug tip. The red microphone wire goes to the plug's middle ring. The bare shielded microphone wire goes to the plug's base and shield. See http://www.hobby-hour.com/electronic...microphone.php for more information on wiring a microphone plug for a PC sound card.
https://www.miataturbo.net/attachmen...ine=1321038673

Next, place the microphone in the battery clip and crimp the clip handle's ears around the plug's shield pigtail. This provides electronic shielding (grounded to your engine block) to reduce unwanted noise from things like your ignition system. It also provides strain relief.
https://www.miataturbo.net/attachmen...ine=1321038673

Finally, pot the heck out of it with hot glue. You're doing a couple of things here: (a) strain relief/robustness/waterproofness/etc. and (b) helping the microphone to only pickup vibrations from the clip rather than the external environment. Note that because of the hot glue and the vinyl covers on the clip handles, this assembly should only be used on the cold side of the engine.
https://www.miataturbo.net/attachmen...ine=1321038673

Finished microphone/plug/clip assembly:
https://www.miataturbo.net/attachmen...ine=1321038673

Assembly plugged into a high-quality 20' 1/8" shielded stereo extension cable (PN 420-2562 -- $12.99):
https://www.miataturbo.net/attachmen...ine=1321038673

Test run. Clipped to the engine mount bolt that secures the starter bracket. Note starter (left), oil filter (upper right) and intake manifold brace (far right).
https://www.miataturbo.net/attachmen...ine=1321038673

The sound quality from this assembly seemed good on the initial test (engine started and idled with some revving). I could clearly hear the hydraulic lifters operating the valves. There was little, if any, hiss or external noise. Volume was good and not distorted. This should work well. I'll let you know after the dyno session on Monday.

You should try the same with an eyebolt, under the head of that same bolt. Securing the mic might be a little harder.

Or, a ring terminal could be secured under the bolt and be adapted to secure the pigtail. Or, you could hack up the battery clip shown, flatten one end and drill a hole for the bolt. If you wanted to build this and leave it in place, that would be cool and would probably improve the sound quality.

Hmmmm . . . .

i used a 2Ga copper lug so I could bolt mine in the factory knock sensor port, so far the hightemp hot glue I used has held but I dont drive around with it on there for very long

Sub'd for results! Perfect timing too! I just decided today to look into making a det can!

Went over this AM and the dyno was broken. Two weeks out.

WTF happened to it? Regardless that sucks! I was really looking forward to hearing results of these... :vash:

Something was wrong with the serial data from the electronic control box. They're sending it off for repairs.

I've been driving around with the det cans. They work great.

Awesome, I think im gonna take the plunge on these.

Sub'd for info. Really want to hard mount a mic to the block using a plate/eye bolt/whetever and run the wires to the stereo plug in the back of my head unit for in car engine noise and the option to plug my headphones into the deck at will. Seems like a good idea for those "did the engine just make a funny noise?" moments.

Nice project Esquire Hornetball.

TorqueZombie, Sean did exactly what you did- plug Jeff's knockbox into his stereo. That way he could listen to his engine while he drove his engine.

Just keep in mind that a 600Ω line-level stereo audio input and an electret microphone jack are two different things -- even though they take the same plug. The electret microphone signal is (obviously) mono. The reason that you use a stereo plug is the microphone jack in a PC provides ~5VDC on the second channel (the "ring") to power the microphone. I doubt that your head unit has a microphone jack, but if it does, you're golden.

Thank you Sir Fae.

Fae's info is spot on. While you can't directly plug the electret microphone into the line-level audio jack of the head unit, you can plug in the output of a microphone power/amp circuit (Jeff's knockbox). There are also fairly inexpensive devices called "Personal Listeners" that can be hacked up for use. They provide microphone power and amplification and have headset outputs that can be plugged into the head unit. I think there are some really cheap ones on E-bay. Radio Shack has one that costs about $30 IIRC.

I just reread this thread. didnt see any mention of what software you used if any to listen to the microphone input live through your headphones. Lemme know what you got! meanwhile im gonna search for something like this.

EDIT: I found an informal way to test the mic, sounds pretty good! Still wondering if you have a better solution than to go into a mic test mode.

In Windoze, I double click on the speaker icon in the taskbar, unmute the microphone input and turn it up all the way. Done.

Thas what I was dooin, just making sure you hadnt found a better solution.

Audacity: http://audacity.sourceforge.net/

Remote mounting the mic in a "personal listener", into the big alligator jack in the pic, is exactly what I did.

I use in-ear headphones to keep ambient noise out. High quality ear-phones help a lot.

Great post. Hornetball thanks for all your help. When ever I post up you alway add something of real value and I appreciate it.

You're welcome.

Subscribed. I assume that potting in JBWeld would be more ideal for a permanent installation.

I think so. Can't think of a downside to J.B. Weld. It is also much more solid, so it should transfer vibration better.

I'm about to build another set of these using a battery lug for permanent attachment to the block. I drive around with the battery clamp version, but I always worry that it might come off the car -- it's probably better suited for dyno use and use as an electronic stethoscope. I think I'll use J.B. Weld on the next set. Will let everyone know how it works. I would expect that a battery lug version with J.B. Weld would yield even better audio.

I've found the following to make things a bit easier when using this setup:
1. Enable sound during recording in Audacity by going to Transport - > Software Playthrough. Make sure there is a check mark on the left.
2. A really good example of what to expect with knock:

Electronic Det Can -- Redux
 

So, I've been using my battery clip det cans for tuning, and they work well. However, there's always a risk that something could come off the car at high speed with these. It's probably just a mental risk (it hasn't ever happened), but I'm still paranoid.

Early in the thread, Curly suggested that rigging this up with some kind of eyebolt setup for permanent installation would work well. Then MD323 suggested a copper lug. Both suggestions sounded good to me, so this is what I built today:

All of the materials are the same as before with the exception that I'm using a #1/0 AWG 3/8" Copper stud ring:
https://www.miataturbo.net/attachmen...ine=1330899292

The microphone element fits perfectly into the lug. Get it in all the way and secure it with JB Weld, high-temp RTV or similar (see post #31):
https://www.miataturbo.net/attachmen...ine=1330899292

Put the plug shields over the microphone wire. You want it to be inside the lug just a bit as we're going to crimp the lug over the shield:
https://www.miataturbo.net/attachmen...ine=1330899292

Crimping away. Crimp enough, but not too much. The wires running within the shield are delicate. The crimp also secures the microphone element within the lug This theory didn't stand the test of time -- make sure the microphone element has been secured within the lug with JB Weld, high-temp RTV or similar -- see post #31:
https://www.miataturbo.net/attachmen...ine=1330899292

Finished assembly with heat shrink tubing making things weather and dirt-proof (I didn't use any other type of sealant or epoxy):
https://www.miataturbo.net/attachmen...ine=1330899292

So, now I have two det can microphones -- one I can move around to troubleshoot car noises, and the other will be permanently mounted for tuning duty:
https://www.miataturbo.net/attachmen...ine=1330899292

3/8" hole works perfectly with the 14mm engine mount bolt:
https://www.miataturbo.net/attachmen...ine=1330899292

Here's the assembly in its new permanent home:
https://www.miataturbo.net/attachmen...ine=1330899292

Really happy with this. The sound quality from the clip-on det can microphone was pretty good, but this is WAY better. I wasn't expecting a sound quality improvement -- but I'll take it.

YMMV.

That looks pretty good. Wondering if you get any better results with the copper lug? You didn't say either way, so I imagine not. I get really nasty high frequency noise and I used all of the exact same materials from the original setup. Anyone else have this problem or any recommended fixes? Is there a way to suppress noise with Audacity realtime? I played around with it for a while and found nothing for realtime noise filtering. Any help would be appreciated.

Yes, the sound from the copper lug is better.

High frequency noise is usually related to electronic interference. Your alternator is an especially effective generator of this type of noise (as are your coils, injectors, etc.). Make sure that the cable you use between the det can microphone and your computer is a high-quality shielded type, and try to route away from the alternator. The use of a shielded plug and good electrical contact between the plug shield and engine ground is really important.

Drove home with the det can. I'm definitely picking up alternator whine like sccaax mentioned. The whine is in the shield, not in the signal, which means I can turn the volume up and have a decent S/N (probably why I never really noticed it before). I think if there was a way to ground the shield at the computer end it would help.

Very nice. So you decided not to go with the JB weld inside the copper lug. Any reason? Or just didn't bother.

I'm working on rigging up an electronic Stand-Alone type today. If all goes well I'll be finished before work but hopefully I'll have pictures up between this afternoon and Monday. This should be perfect to monitor my 1.6 pre and post-turbo, and should work better for me since I feel more comfortable listening actively through headphones rather than through my computer or a recording. This has been a very inspiring thread so far (I hate working on electronics/wiring yet I'm doing this anyway)

With the crimp, didn't see that it was needed. I don't think it would hurt anything though. Would be a little messy.

I'm going by the Shack today to pick up some Ferrite Clamps and see if I can cut down on the alternator whine in the cable shield. Will let you know if that works. Any EE's out there have any advice for the alternator whine?

Fixing the Fails
 

Fail #1, Alternator and Ignition Noise:

Scaax has this bad. I definitely had some too. Investigating, I tried the following:

1. Ferrite clamps: Made no difference at all.
2. Plugging the laptop into the cigarette lighter to give it a ground. Made things a LOT worse. I think I actually set up a ground loop that exacerbated the noise.
3. Ditched the Radio Shack "shielded" cable and fabbed my own. WIN -- no noise whatsoever -- problem solved.

Let's look at why #3 fixed the problem. The Radio Shack cable is two conductors with a shield. The shield acts like an antenna and picks up RF energy from the environment (mainly the alternator). The shield also doubles as the ground wire for the microphone assembly and is terminated at both ends (i.e., that alternator noise goes directly into your computer audio card ground plane). What we need to do is route the alternator whine into the engine block and keep it away from the computer audio card.

To accomplish this, I sourced cable that had three conductors and a shield. The three conductors are terminated to a 1/8" stereo jack (Radio Shack P/N 274-274) on the engine block end and a 1/8" stereo plug (same as the plug used for the microphone assembly) on the computer end. The shield is joined to the ground conductor ONLY ON THE ENGINE BLOCK END. DO NOT TERMINATE THE SHIELD ON THE COMPUTER END. I had some really high-quality aircraft signal cable laying around and used it to build my new cable. This cable had three twisted conductors, a full-length braided shield and Tefzel covering. Overkill, but it worked great.

So, to summarize, build your own cable, terminate the shield of your cable ONLY AT THE ENGINE END, and make sure you don't power your computer from the car's electrical system (have it run off of its internal battery). Do this, and you will not hear unwanted noise.

Fail #2:

I'm pretty sure my microphone is loose and rattling a tiny bit inside the copper lug. I get some noise that sounds like static or an overdriven speaker when I sit in the garage and rev the engine. This wasn't happening when I first built the lug microphone, but it has developed over the last few days as I've driven the car. Unfortunately, the noise sounds a bit like det, so it is definitely something that needs to be addressed. I'm wishing I had secured the microphone with JB Weld or high-temp RTV. Looks like I get to build another lug microphone . . . practice makes perfect?

I had my suspicions about that cable when I bought it. But I guess it breaks down to how it's used. Would this work? - Get another male plug and a female plug and solder the first two rings together for the plugs on the computer end leaving the shield unconnected at the computer. That way, I can reuse the cable I've got and its a bit cheaper as well.

I would recommend Gorilla glue as a higher temperature long-term solution. Working really well for my battery clamp so far. It bubbled like crazy. Someone recommended baking the air out, but I didn't want to risk damaging the mic. I don't see a problem with the bubbles. Heck, they're a great insulator and nobody ever sees it anyways. I do like the copper lug idea.

I don't think that will work because you still need a ground wire to interconnect the ground plane of the computer's sound card and the (-) end of the microphone.

Here's a link with some info:

http://www.hobby-hour.com/electronic...microphone.php

Building a cable isn't much more money than you're talking about anyways. All I did was get a length of 3 conductor shielded cable, a male plug and a female jack.

Also, if you're still using the battery clamp version, take a look at how I connected it in the original write-up. You'll notice the cable's initial routing was down and away from the alternator. There is less noise if you do it that way and try to use engine structure to block the alternator. Better cable is the right solution though.

More good info. Thanks guys. Do you just run the cable into the cabin through an open window?

Edit: Just ordered some of that cable from Aircraft Spruce since it's not terribly expensive.

You can actually run it over the seal between the firewall and hood and just under the quarter panel and through the door near the hinge. That's what I'm doing for a temp setup. A few less looks from the bacon.

Hornet, I thought I understood what you were doing but now I'm more confused. The - from the mic goes to #3, then how are you talking about grounding the shield any differently? Can you sketch up something quick in paint? The link doesn't help about the shield bit. Is the shield essentially another path on the wire? How does it interface physically on the connector(s)?

Ah, I think I get it now: The radioshack shielded cable is really 2 conductor and the shield and neg are both carried on the jacket and at Pin 3 on both ends. With that said, couldn't you just make sure the neg/ground doesn't touch the lug or battery clamp and rely on the neg on the audio card end? I would think that this gets rid of any possibility of completing the loop by taking the same ground as the alternator and any other non-solid-state component out of the mix. Provided the ground at the PC isn't too noisy? This is an ME talking not EE. Help me out.

Well, I'm an AE (x2) and JD. But I do work in avionics.

The RF from the alternator is powerful and has to go somewhere. It's not a ground loop issue. You've got to give that energy an outlet, and the proper outlet is into your engine block. If you let it into your computer, you'll hear it. You'll notice that the alternator noise is constant -- it doesn't change volume with the slider. In fact, you can move your slider all the way to the minimum position, and the alternator whine will still be loud and clear. That's a dead giveaway.

So, to summarize:

Three conductors and a shield, let's call them:
1. MIC+
2. MIC-
3. MIC Bias (3-5VDC from your sound card)
4. Shield.

On the jack at the det can end, wire as follows:
1. MIC+ to tip.
2. MIC- to base.
3. MIC Bias to ring.
4. Shield to base.

On the plug at the computer end, wire as follows:
1. MIC+ to tip.
2. MIC- to base.
3. MIC Bias to ring.
4. Shield is NOT terminated.

While it is true that at the jack, Shield and MIC- are wired together, the alternator whine will seek the easy exit through the engine block rather than follow a smallish 22-gauge wire for several feet into your computer. That's why this works.

Profit.

Just built one of these. Thank you hornetball for the awesome write-up. I potted mine in JB weld and used all the parts mentioned in this thread except for the 3/8 lug, since the mic element wouldn't fit in the lug I bought at walmart. Since I was working on this at midnight and I was desperate to get it done, I found out that the EGR pipe from a 95 motor is the perfect inner diameter for the mic element lol. Crushed one end flat and drilled a 10mm hole in it, and will be bolting it up to the stock knock sensor location on my 2001 block.

Note to anyone else doing this... I wasn't thinking the first time around, and forgot that a microphone works because of a moving diaphragm. I just had in my mind that fully surrounding it with JB weld would give it rigid contact at all points, allowing it to pick up more sound. If you fully surround the mic element with JB weld, though, you can potentially limit the movement of the diaphragm and get really shitty results. This happened on my first attempt. Thankfully I tested the wired-up mic element before potting it, otherwise I wouldn't have realized how much the noise level decreased after potting.

For the first one, I put some JB weld in the tube, inserted the mic, and then filled it up the rest of the way with JB. It worked, but the sound was ---- compared to before potting. For the second one, I just inserted the mic element into the empty tube, gave it a light press with a dental tool to be sure it was at the bottom of the hole, and then filled the rest of the tube with JB weld, leaving the front face of the mic element exposed. I've just bench tested it so far, but sound is MUCH better on the second version. At least as good as the bare mic element, if not better. Bolting it up to the motor tonight, hopefully doing some tuning this weekend.

Having an issue with mine. I can get great sound for a second or two, but what usually happens is that when I hit the listen/record button, I get a super loud blaring sound for a couple seconds, then a couple seconds of perfect engine noise, and then it fades down through a few different sounds until I get nothing, or sometimes very faint engine noise. Basically, the sound level continually decreases from the time I turn it on until nothing.

I've tried the standard windows 7 microphone listener, the Dell sound controller on my laptop, and Audacity, and all give slightly different but similar results. Audacity gives me a useable recording if I just let it keep recording after the volume in my headphones drops off while listening live, but that still doesn't let me listen in real time.

Ideas? Software or hardware? Starting to seem like hardware to me...

The fact that you can get a usable recording indicates that most things are basically working. Does your headset incorporate an amplifier? When you listen to the recordings, are you using the same computer and headset?

Yep, everything has been done on the same dell laptop, including listening to recordings. I can't imagine the headphones would have a built in amp... they're just some cheap earbuds that I carry for music listening.

Hey! How are you guys using your lap tops as a listening device to listen to Det in real time.
I have an old pc mike with a long Coax wire that should be great for this.

Thanks

Well, I just plug the mic into the mic port and turn up my volume (make sure the mic input is enabled in "Volume Control") . Using a headset is highly recommended.

Cool thanks. I'll try again.

I must be stupid or my thing is broken :vash:

Wrong thread.

This thread coming back got me to thinking about if anyone has used a non electronic version of det cans? Copper lug or a piece of copper tubing then to a hose like air compressor hose into the car. The gray foam water pipe insulation for additional sound deadening from outside sources. To a stethoscope (like a doctor uses). I just wondered if anybody has tried both. I made an electronic version and this version and felt that I could hear the beginnings of knock better with the manual version. It was more difficult to install but the sound transferred into the stethoscope really well.

I built one of these but seem to be picking up a lot of external noise with it -- intake, BOV, etc. If you have a recording of how it's SUPPOSED to sound, can you attach it?

Dude, it's a mic. It's designed to pick up sound.

The lug version will be better at picking up engine block sounds vs. external sounds than the clip version.

It's picking up so much I doubt I could hear detonation over the racket. I'm building the clip version today.

T H A N K Y O U
for the audio example

Bump because I'm going to build one too, but have a question.

The link you provide shows the tip and ring get connected together to one of the microphone wires, and the second microphone wire gets connected to the shield and ground. So which is it?

Maybe this will become more obvious when I find the microphone; the tiny RadioShack in the mall didn't have jackshit.

You'll have to look at the packaging for the microphone. On the ones I can find around here, the microphone had the 3 wires (red, white and bare) and the packaging described what each wire's function was.

RadioShack is a shell of what it used to be. They are just another wannabe cell phone retailer these days and may not last much longer. There's always Mouser.

Just a bump and some questions because I just bought a set of those noise canceling headphones (for range use) that also have an audio in jack.


Basically I got the above in black, for $49 at Home Depot. Such a deal.

I am wanting to use these as my headset for listening when not plugged into a laptop. I saw "personal listener" described above but it seems like the versions I am finding on ebay tend to be the ones that are designed as a serious assistive device and are quite pricey.

Anyone have any search terms or sample listings I can check out? Is creating a DIY mic power box something that would be easy to do?

Those looks like those might have a built in amp already, maybe they dont need an external amp to work.

Searching for "headphone amplifier kit" on Amazon has turned up some results. is prebuilt, about $36 bucks, and looks reasonably durable.

Personal Sound Amplifier

I use this, extended the microphone to the engine and left the "amp" in the center console. I just use earbuds, but you could use those with a M/M 3.5mm cable. You can get these in store, or at least I did.

Driving around with headphones isn't exactly legal in some places, I prefer the earbuds for discretion.

sorry for digging up
but can this be used with the donnut style knock sensor?
maybe for better sound quality?