Beesneez Lulu Fet - KM84 SMD mod

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
A few years ago, i troubleshot the original mainboard of my Sharp TV by removing & resoldering 0603 and 0805 ferrite beads with a soldering gun (probably almost as old as me), with the tip made out of a ~1mm diameter copper wire loop ;D

Where there's a will, there's a way... ;)
 
Khron said:
That soldering technique reminded me of AvE ;D https://www.youtube.com/watch?v=-vhHYo8WOFk

I'm intrigued about the solder specs for those 1G resistors. Got a link to the datasheet, or wherever that's mentioned?

LOLz. That was a pretty funny watch! Nice style of presentation.

Here's the datasheet for the 1G: http://www.mouser.com/ds/2/303/res_hvf-472024.pdf

I think I know what you're thinking... "Why is it bubbled underneath the package?!"

The answer: "I'm not really sure". I assumed it would be fine with the reflow, but I might have been wrong. I thought it had info on that on the datasheet, but I couldn't find it when I just looked through it... :/

Any thoughts?
 
Khron said:
That soldering technique reminded me of AvE ;D https://www.youtube.com/watch?v=-vhHYo8WOFk

I'm intrigued about the solder specs for those 1G resistors. Got a link to the datasheet, or wherever that's mentioned?

Not every Canadian is that silly,  by the way. Just most of us.
 
Part Six - Big Bits

There are only three main "big bits" on these mic boards, the XLR connector, the transformer, and the "SDC Pointybit™" (see: https://groupdiy.com/index.php?topic=65883.msg834676#msg834676).

It took me while to find out exactly where I could get the connector, a male plug insides, but it turned out Switchcraft were the people to talk to. The part number is: QG3MBAU for the gold plated connectors if anyone needs to get a hold of some.

I trimmed the ends down with a rotary tool and got them soldered on the board. It's worth noting that pins 1 & 2 are soldered to the top of the board, and pin 3 is soldered to the bottom so when I designed the board I made sure I poked the pads on the correct sides in the Eagle component package. It's just a pity I covered them with solder resist!! Because the pins are not only a little long, but too wide to slot the board in-between, I needed to widen the space between the top and bottom pins to accommodate them. Once that was milled out it was easy to solder it on to the board.

There were a couple of things I wanted to be doubly sure on, 1: That the connector was soldered dead straight (otherwise it's a bit awkwardly tight when you go to plug in your cable), and 2: That I didn't melt the plastic holding the pins in place while I was soldering (or you have the same pain-in-the-arse problem when you go to connect your cable). It didn't take too much to get it right, but I felt like spending the time thinking through it would be worth it. My cables connect nicely, so it must've been worth it. Yay!!

The transformer was fairly straightforward to find/decide upon. The Lulus used a Cinemag CM-24110 4:1 but I decided to try the CM-5722 6.9:1 instead as it was modelled on the original KM84's BV-107. In hindsight, I think I could've stuck with the CM-24110 and been perfectly fine. Obviously it was a great choice to put it in the Lulu, or Ben wouldn't have chosen it in the first place, but I thought it might just be fun to try out something close to the original.
 
Was it all worth it?

Hmm, probably not so much, as far as I can see. Maybe I could ground the copper foil in the CM-5722 in a future version (I chose to trim it off). Maybe I could set up a 50Ω-200Ω impedance change jumper, similar to the KM84… But who needs to change the mic impedance these days? To be fair, the CM-24110  is slightly cheaper and, if I've understood turns ratios, gives a little jump in output so I might stick with that in subsequent builds.

Getting it into the mic was fine once I'd milled out the space a little more. I chose to have a reasonably tight fit so I don't feel any transformer movement when I move the mic around. If I wanted a little more security I guess could use a rubber gasket and a slightly larger hole, but I think I'd end up with fairly narrow sides on the PCB and risk buggering up the traces running along the side.

Another small, but worthwhile, decision was to make sure the trafo leads were nice and short. I didn't want to have any extra flying around inside. Plus, the leads could all be soldered to the rear of the board leaving the top looking rather clean and tidy. I've been thinking in a future layout I might make use some sort of connector pins to tidy it up even more.

"My momma always said 'tidy is as tidy does'"…

Finally the "SDC Pointybit™" came from the Lulu board. I couldn't find an aftermarket part so rather than make one I just nicked the part from the original board. I'm pretty sure I know how I'd do it in future, but I couldn't be bothered reinventing the wheel for that part.

With those Big Bits attached I finally had my mics ready to test!!
 

Attachments

  • Two KM84 SMD and Lulus - Complete 1.jpg
    Two KM84 SMD and Lulus - Complete 1.jpg
    1.6 MB
And.... how'd they sound?


If you want a very cheap mic that you can bring up to spec these are pretty good for the money

http://pages.ebay.com/link/?nav=item.view&alt=web&id=311911932431&globalID=EBAY-GB
 
They sound great, thanks Timjag!

Well, they both sound great, but I've an intermittent issue with No. 2 where it will more often than not be incredibly quiet, like 20-25dB down, but at other times be perfect. Nothing changes in the tonality, just the gain. I'm not, by ANY stretch of the imagination, an electronics wiz... Which probably makes this seem like a bizarre project to most people... But, on the odd opportunities I've had to work on the poor old girl I still haven't sound the solution.

I've been wondering if populating a whole new board and swapping it out might be the best solution rather than trying to track down the problem.

That's probably one of the slightly harder things about using SMDs instead of thru-hole. :/

So, as far as sound goes: To my ear, for whatever that's worth, they sound really similar to the original circuit just a little quieter (and flat/linear) noise floor. Which is exactly what I was aiming for, so YAY! The capsules in these mics are brilliant, they're made by Ben Sneesby so that's a given, but to negligibly change the character while lowering the noise floor even more is a great result from my point of view.

Certainly a good result given that I'm not an electronics engineer and had to learn almost everything from scratch to give this mod a go!

Ha, noobs FTW.

ps. thanks for the link, good price on those mics alright. Might be worth a check out!
 
25fb is a lot, and I take it you're sure it's not the desk preamp? A loss like that can be caused from going from balanced to non balanced, can't remember but I think the 84 is a single fet? So it sounds to me like a connective issue. But SMDs are a bugger being so small - check the transformer connections etc.

Yeah those little mics are great the capsule is surprisingly ok and you get the pokey springy bit inside to connect to the capsule, I I stuck a c61 nuvistor circuit in it for S&G and it came out sounding nice. I mean for $100 you can't really grumble for a donor SDC.
 
If we're talking cheap SDC donors, what about the t.Bone SC140's? You can get a stereo pair (including shockmounts and flightcase) for 89eu ;D

From what i've understood, they are, at least mechanically, Rode NT55 knock-offs (as in, Rode capsules do fit onto the SC140 bodies).

I have two pairs of those; i made a PCB for one with an Oktava MK012 circuit (and "transplanted" the DC-DC converter from the stock board), but i think i'll end up KM84-izing all four of them... one of these days :p
 
Yeah, that KM84 circuit is pretty straight forward even for a noob like me! It might not be much help for pros like you guys but I think I posted the Eagle file for mine earlier if you want one ready to go, if it would fit in those mics.  :)

I was reading the other day about negative feedback to lower the noise floor even more. What do you guys know about this, is it worth trying to implement it into the KM84 circuit to improve the signal to noise?
 
Ok, I've measured the voltages throughout the circuit and there's a few differences between what's prescribed and what's there.

I seen that under load it'll be different, but I'm a little perplexed at the 46V at R1 vs the 0.5V that I'm measuring. The two mics have slightly different voltages the closer it gets to C2. Mic 1 is running well, but Mic 2 is the one that's low on output.

Any thoughts?
 

Attachments

  • Circuit wt Voltage.jpg
    Circuit wt Voltage.jpg
    275.2 KB
What's the value of R3?
If it's the same as in original KM84, then you should change it to lower value and set bias properly.
You should get higher output and probably easier proper setting.
R1, R8, R5 - high impedane section - don't expect that you will measure proper values wth typial multimeter which usual have 10M input impedance.  If you have extra 1G (THT) resistor then connect it in series with positive lead of multimeter and change range to mV. Then your reading will be much more accurate.
 
R3 is a trimmer in my mics so I've set that do get the lowest 2nd harmonic I can following Karl Adams' YT video: https://youtu.be/SHD4Rrwqy_Q

I assume I've done this correctly, though like everything else I'm still pretty new to all this sort of fun so let me know if there's something else I should be looking at. ;)

I'll attach a screenshot of what the signal looks like. There's a bump in the low end, but that's just them humming.
 

Attachments

  • Both Mics wt Signal.png
    Both Mics wt Signal.png
    1.1 MB
Value of your trimpot have matter according to fet type. If it's 10k-20k then i would use lower probably.
 
Back
Top