4ch Balanced mike splitter with jensen transformers inside the ART P16 patch bay

Shop deals on ebay
Advertise with us
shop deals on amazon
GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
M

Making_Waves

Member
Joined
Nov 1, 2012
Messages
10
Hey everyone, I've never posted on here before, but a friend suggested I should post my most recent project here.

I made a 4 channel balanced, passive microphone splitter based on the Jensen application note 005. For the enclosure, I gutted the ART P16 XLR patch bay, created my own pcb, and threw it in the enclosure.

I decided to make it because there didn't seem to be many options with passive mike splitters on the market. I chose to use the ART P16 patch bay to save a lot of time, money and effort by not having to search for the right sized enclosure (the P16 is not very deep for a 1 rack space unit), not having to buy XLR connections, and not having to drill the holes for the XLR connections.

It can be easily expanded to 8 channels, just use 2 PCBs, and double the quantities of parts. The following part list is listed for only 4 channels.

Cost

Mouser part numbers are given for everything but the transformers

Jensen JT-MB-CPC  x4 - about $230

ART P16 XLR patch bay - I got mine used from guitar center for about $70

Rocker switch (612-RD1113112R) x4 - $2.68

51 ohm resistor (279-LR2F51R) x8 - $.88

2.7k resistor (603-MFR-25FTE52-2K7) x4 - $.48

.01uf capacitors (505-FKS2D021001B00JI) - $1.44

circuit boards - $35 for 5

Total for four channels: About $330

For 8 channels, you wouldn't have to buy more PCBs, or another enclosure, just more transformers and components so for 8 channels, it would be about $560

Here's a dropbox link to my gerber files for the PCB:
https://www.dropbox.com/s/1uvnrk0sq5tm93m/Jensen%20Mike%20Splitter%20v2.zip?dl=0

And you can find some pictures here:
https://imgur.com/a/XcwOE

Schematic can be found on Jensen's website, application note 005.

Clearly, you can mix and match some different components to get the price down. Most obviously, the transformers. I went with Jensen because I didn't want to cut any corners. Also, if I could do it again, I'd choose a different way of approaching the ground lift switch. I had to dremel out a little cut out for each of the switches, and create short jumpers to each tab with some slack so I could move them into place.

So yeah that's it! Again, this is my first post so I'm not sure if I'm missing any info or anything let me know!
 
Making_Waves said:
Hey everyone, I've never posted on here before, but a friend suggested I should post my most recent project here.

I made a 4 channel balanced, passive microphone splitter based on the Jensen application note 005. For the enclosure, I gutted the ART P16 XLR patch bay, created my own pcb, and threw it in the enclosure.

I decided to make it because there didn't seem to be many options with passive mike splitters on the market. I chose to use the ART P16 patch bay to save a lot of time, money and effort by not having to search for the right sized enclosure (the P16 is not very deep for a 1 rack space unit), not having to buy XLR connections, and not having to drill the holes for the XLR connections.

It can be easily expanded to 8 channels, just use 2 PCBs, and double the quantities of parts. The following part list is listed for only 4 channels.

Cost

Mouser part numbers are given for everything but the transformers

Jensen JT-MB-CPC x4 - about $230

ART P16 XLR patch bay - I got mine used from guitar center for about $70

Rocker switch (612-RD1113112R) x4 - $2.68

51 ohm resistor (279-LR2F51R) x8 - $.88

2.7k resistor (603-MFR-25FTE52-2K7) x4 - $.48

.01uf capacitors (505-FKS2D021001B00JI) - $1.44

circuit boards - $35 for 5

Total for four channels: About $330

For 8 channels, you wouldn't have to buy more PCBs, or another enclosure, just more transformers and components so for 8 channels, it would be about $560

Here's a dropbox link to my gerber files for the PCB:
https://www.dropbox.com/s/1uvnrk0sq5tm93m/Jensen Mike Splitter v2.zip?dl=0
And you can find some pictures here:


Schematic can be found on Jensen's website, application note 005.

Clearly, you can mix and match some different components to get the price down. Most obviously, the transformers. I went with Jensen because I didn't want to cut any corners. Also, if I could do it again, I'd choose a different way of approaching the ground lift switch. I had to dremel out a little cut out for each of the switches, and create short jumpers to each tab with some slack so I could move them into place.

So yeah that's it! Again, this is my first post so I'm not sure if I'm missing any info or anything let me know!
Click to expand...

Because you certainly have the room on your Mic-Splitter PCB, I would have used much wider routing tracks than what I see in your photo. Using thin routing tracks not only increases the impedance, but also the inductance. I would have used 25-mil, 30-mil or even 50-mil routing tracks and then "necked-down" the track width when terminating into a component pad if necessary. Just my 2-cents.....
 
MidnightArrakis said:
Because you certainly have the room on your Mic-Splitter PCB, I would have used much wider routing tracks than what I see in your photo. Using thin routing tracks not only increases the impedance, but also the inductance.
Click to expand...
The resistance and inductance of a 6 mil PCB trace is way smaller than the DC resistance and leakage inductance of this transformer.
I agree that, for mechanical purpose, the pads should be large.
 
Hey! After three years I finally get some feedback! Thank you so much, this is the first pcb project I did and have since started working on an electrical engineering degree to learn how to design other audio gear. I've also updated this circuit to include a small, phantom powered, in-line amplifier (mic activator) on the transformer isolated output to help accommodate for the -3db reduction in gain when splitting. I think I posted in another thread on here about the design. You can find it in my post history. I still get lots of use out of them.
Thanks again for the feedback though. I had no idea that thin traces can lead to more inductance and impedance.
 
Making_Waves said:
Hey! After three years I finally get some feedback! Thank you so much, this is the first pcb project I did and have since started working on an electrical engineering degree to learn how to design other audio gear. I've also updated this circuit to include a small, phantom powered, in-line amplifier (mic activator) on the transformer isolated output to help accommodate for the -3db reduction in gain when splitting.
Click to expand...
Have you actually measured this 3dB attenuation? It should be very much dependant on the mic impedance and teh input impedance of the various loads.
Making_Waves said:
I think I posted in another thread on here about the design. You can find it in my post history.
Click to expand...
Do you have a link?
Making_Waves said:
I had no idea that thin traces can lead to more inductance and impedance.
Click to expand...
As I wrote earlier, this is so unsignificant you can forget about it. The resistance of a 12mil 4" long 1oz trace is 0.16 ohm. Compare to a microphones 150r nominal impedance, and particularly to the input impedance of a mic pre (typically 1-2 kohm). Or compare to the DC resistance of a Jensen xfmr.
 
abbey road d enfer said:
Have you actually measured this 3dB attenuation? It should be very much dependant on the mic impedance and teh input impedance of the various loads.
Click to expand...
No I haven't measured this, but in theory, I'm splitting the microphone signal to two preamps that should have similar input impedances and therefore there's be unavoidable signal loss that's around -3db.


abbey road d enfer said:
Do you have a link?
Click to expand...

Sure! Here's my post on the thread, and it includes some pictures of the populated splitter pcb:
https://groupdiy.com/threads/active-ribbon-mic.5743/page-17#post-941362
I reference PMM's schematic on page 5 here:
https://groupdiy.com/threads/active-ribbon-mic.5743/post-73762
And I used this post to decide on the transistors to use:
https://groupdiy.com/threads/active-ribbon-mic.5743/post-840596
It's a very simple fixed amplifier with low gain (I haven't measured how much exactly) and I stuck it into the transformer isolated part of my original circuit. I found that it's really helped me when recording quiet sources in live environments. If money was no object, I could use 16 clean, high-gain pre-amps for every signal I record and not need these little boosters, but unfortunately that's not the case for me.

I've actually updated the circuit once more since that post to incorporate a toggle switch so that the "boost" can be bypassed.
 
abbey road d enfer said:
I would bet the attenuation is closer to 1dB.
Click to expand...
If I have a voltage source driving two loads, wouldn't I measure half the voltage across each load when compared to one load? Then half the voltage would be -3db. Of course this is assuming the two loads are relatively close in impedance. Am I missing something major here?
 
Making_Waves said:
If I have a voltage source driving two loads, wouldn't I measure half the voltage across each load when compared to one load? Then half the voltage would be -3db. Of course this is assuming the two loads are relatively close in impedance. Am I missing something major here?
Click to expand...
You're missing the fact that we are not in a world of matching impedances, rather bridging, where the load impedance is much klrger than the source impedance.
Typically, a 200r mic meeting a 2k mic input results in about 1dB attenuation.
A 200r mic meeting two mic inputs (equivalent Zin 1k) results in 2dB attenuation. So the difference is about 1dB.
 
Making_Waves said:
No I haven't measured this, but in theory, I'm splitting the microphone signal to two preamps that should have similar input impedances and therefore there's be unavoidable signal loss that's around -3db.
Click to expand...
You're thinking in the context of matched-impedance sources and loads - it doesn't work that way when source is low-Z and load(s) are high-Z. Consider a slightly more extreme case of 100 Ω source and two 2 kΩ loads. In parallel, those two loads look like 1 kΩ. Total signal ("insertion") loss with a load ten times the source impedance is only about 1 dB.
 
Incidentally, you should consider the 51 Ω resistors shown in the Jensen AN-005 optional. When I wrote that, I'd included them - as a means to damp possible standing-wave issues in the unintended transmission lines formed by mic cables and the "earthy" environment (as described briefly by Ralph Morrison in one of his excellent books). After years of hearing about real-world customer experiences, inclusion of the resistors has been neither better or worse than without them. The more important consideration is to keep the leads of the capacitors as short as practically possible (to keep the capacitors low-impedance to the highest possible frequency). Thanks for using the Jensen transformers - you won't ever regret it. As the late Deane Jensen (company founder in 1971) used to say, "When you buy a Jensen, it only hurts once - later, their performance will make you glad you did." He left the company to me in his will when he died in 1988 - and I carried on his work until I sold the company in 2014 and "semi-retired."
 
Actually, the 51r resistors have other purposes.
They provide protection against short-circuits, which happen quite often in live recording circumstances. Without them, a short reflects at the primary as a very heavy load that the driver has generally difficulties in handling. Even if the driver continues to deliver some signal, it does it with severe distortion and risks of polluting the power rails. What's more the other output(s) are severely attenuated.
Marginally, they also provide adequate damping for some transformer-based mic preamps, that exhibit significant resonance when driven by a very low Z.
 
The resistors I'm talking about are the ones in the "grounding" path for XLR pin ! to chassis (as shown in Jensen AN-005). You're talking about resistors in the signal path to reduce the effect of shorts on one output of a splitter.
 

Latest posts

Back
Top