newbie builds ioaudio MK47 microphone

[chunger]

A small update. . . I've mostly been hacking metal to build the rev. 3 PCB prototype.  We still have a few minor tweaks to do on the board, but for the most part, it is very close.  I think if I don't run into any MAJOR malfunctions, we will be able to move to full production PCB while implementing the very small adjustments.

My main purpose for this prototype build is to test function with the lower-cost, smaller, lighter, more readily available Triad 230V 25VA transformer that I had used on my 1st 2 builds on this thread.  We will make sure the final case design fits the 200V Antec with the 6.3V taps because we do have plans to use the same case configuration and standoff locations to implement some of the other PSU designs for various microphones.  I also want to nail down a final BOM for the 230V transformer since many of the values shift when the voltmeter is introduced into the passive PSU topology.  These low-cost Chinese meters do not come with any documentation so it takes a little bit of experimentation to figure out the right resistor values.

I cut a 3" piece of aluminum L-bracket to mount my toroid and choke.  Holes were drilled and tapped to install this L-bracket.  I also cut a hole in the front face for a digital voltmeter as well as a 3/8" hole for the rotary pattern switch that I like to use.  As I was merrily building, I remembered that Matador wanted detailed updates on the prototyping process because he usually does this stuff and doesn't have his workshop set up yet, so I brought out my camera and started properly documenting.

One of the most critical PCB positioning considerations is the attenuator pot which is now directly PCB mounted on the 3rd rev. board.

The end of the pot needs to sit nearly flush with the outside case so a screwdriver can be used to adjust B+ externally, but the screwdriver will not be able to enter the case by accident.

Some may find this method a bit crude, but I clamp and use the actual PCB as a drilling template so my standoff holes are exact.  I know this board will fit with 100% certainty.

Once the holes are drilled and chamfered on the back side for flat head screws, I install my standoffs.

At this point, all of the metal hacking is done, and I can move on to assembly.  If all goes well, Dan Deurloo will be able to make the final cases with pre-installed standoffs and pre-drilled holes so kit builders will essentially start from this point with everything directly bolting together.

I start by wiring up the mains.  Since I am in the US, I wire the two 115V primaries in parallel.  And, the secondaries should be wired in series for all builds to generate 230V.

With a quick test-fit of the PCB, we can get a good picture of how the internal wiring will run in this layout.  I feel it is pretty intuitive and the wire runs are straight forward.  Matador has shifted the main screw terminal to the side of the board for easy access and he has put the test switch on the PCB as well to simplify the build further.

That's all for tonight. . . I feel like the rest should be downhill, and I am hoping to have this up and running quickly.  I have no clue what I'm going to do with all of these mk47 PSU's    I already have 3 built on this thread. . . this one will be a 4th, and I will at minimum have one more build to do for kit instruction documentation 

 

[chunger]

More progress last night. . . PCB is populated.  There really aren't that many components on this thing, and the heatsinks dominate the real estate.

Next, I wire up all of the chassis mounted connectors and components. . . from this point, everything should connect to the PCB with screw terminals.

The PCB fits in with minor obstructions this time around.  We are getting closer to an intuitive final layout 

And, we are getting closer to a nice, clean front face layout as well.  I opted to mount the Neutrik connectors from the rear on this build and chamfer the holes.  Using flat head stainless steel screws, this will make solid chassis contact for shield grounding.

And, the PSU is buttoned up and ready to fire up.

 

[chunger]

Powering the unit up was uneventful which is good.  No smoke! 

Also surprisingly, my voltages adjusted right to 105V with room to spare on both ends of the range.

The meter supply voltage was low though at about 2.8V with a 35.5K/63.2K divider at the LED supply.  This is where we are drawing power for the 3 wire meter.

Moving the split to 27.4K/68.1K yielded 2.9V which is closer, but not enough for our meter that needs between 3-30V on the supply.

Another change to 18.2K/68.1K produced 4V which is in range.  I like sitting on the lower end of the range because if the PSU is turned on without a load, the open supply will spike up to around 275V pulling the LED and meter supply with it.

The LED is a little bit weak still at 1.7V, so maybe some more tweaking will be in order.


The bigger issue in my opinion is the meter itself which is model YB27_v1.7 (red LEDs) .  The meter is a bit "jumpy" not settling very well.  Contributing to the instability of the reading is the tendency for the passive supply voltage to vary depending on how many LED segments are lit.  The previous prototype which used a YB27_v1.6 (blue LEDs) was better in this regard.  This issue aside, there is an attenuator on the meter and it can be precisely calibrated to 105V which is the point where we require accuracy.  Once the LED segments stabilize, things settle down.  Cleaning the meter with some isopropyl alcohol seemed to help the issue, but I am not sure if it was the cleaning or the new meter.  I purchased quite a few samples and tried changing the LED display on my original meter with yellow one  ;D but that was unsuccessful as the pinouts were different.  The reason I switched to the v1.7 meter is I found it in the 0-300V DC range which suits our application.  But, I would prefer the more numerically stable v1.6 (0-200V) under normal use if I can find it again at quantity.

 

[kcatthedog]

David,
You use one pair of blue and green for the lamp. There are 2 6.3VAC secondaries each utilizing a blue and green wire.
DO NOT tie the two greens and two blues together.

The 200VAC you get from using the yellow and white lines. Put some shrink tube on the end of the grey line. It is unused

Dave

Hey Wave,

I am doing the  mk-u47 build using the Black zayance pcb and mouser BOM; just about done.

I have the Triad 553-vpt230-110 toroid with purple/blue going to switch and  grey/brown to IEC and have  the yellow/black going to pcb and orange /red left over .

Chunger's pics show this being shortened, soldered together and insulated off.

Do I have the option of running the lamp and, if so  how ?  I have one of Dan's Collective Cases.



thx !

 

[kcatthedog]

Chunger,

Did I understand you correctly and have you wired the cable for the 47 identically to your c12 build ?

Thx !

 

[wave]

David,
You use one pair of blue and green for the lamp. There are 2 6.3VAC secondaries each utilizing a blue and green wire.
DO NOT tie the two greens and two blues together.

The 200VAC you get from using the yellow and white lines. Put some shrink tube on the end of the grey line. It is unused

Dave

Hey Wave,

I am doing the  mk-u47 build using the Black zayance pcb and mouser BOM; just about done.

I have the Triad 553-vpt230-110 toroid with purple/blue going to switch and  grey/brown to IEC and have  the yellow/black going to pcb and orange /red left over .

Chunger's pics show this being shortened, soldered together and insulated off.

Do I have the option of running the lamp and, if so  how ?  I have one of Dan's Collective Cases.



thx !

You have to tie the red and orange together to get the 230 VAC you need.
There is a provision on the PCB for an LED. This is for your indicator lamp.

Dave

 

[kcatthedog]

Hey Dave,

Thx, yes I have the led for the pcb.  I notice in chunger's bom for his ptp (sorry too many acronyms) he references another mouser part to mount the led in the case ?

So if in addition to the led I want to run a guitar amp jewel light , I can buy the complete houseing and wire it to  the teh red/orange tied together ?  I thought I saw that in pics.

btw if any one is wondering about the pin wiring :

Following the pins it would be colour coded like this;

Pins Mike psu 7 in psu xlr out   Comment
1       white white       white
2       blue blue   blue
3       black black       black
4                                         (ganged at mike pins 3/4/6 so running on black but I could run any other colour than white, blue, red)
5       red red            red
6     (ganged at mike 3/4/6 pins so running on black but I could run any other colour than white, blue, red) and different than 4)
7      shield/Green    =>          Chung did this to tie everything in the psu to star ground, which can help with quietness and certainly with safety

thx to Cat 5 for max's schematic !

 

[wave]

Hey,
I wouldn't wire anything to the orange / red lines. They should just be tied and insulated off.
Are you located where the wall AC is 120V?

Dave

 

[kcatthedog]

Yes,  only  a 19 hours and 3 minutes drive north east of you; Ottawa, Canada !

 

[wave]

I put this in my DU67

https://www.tubesandmore.com/products/P-L124-P

http://www.mouser.com/ProductDetail/Chicago-Miniature/949/?qs=sGAEpiMZZMucm%2f%252bFOY0TQv%2fve8X86S41s8SZFySeIzI%3d

Ties directly to the IEC and power switch...
Here's a pic of my 67 PSU

http://groupdiy.com/index.php?topic=50021.msg638477#msg638477

Dave

 

[kcatthedog]

Thx Dave, parts ordered.

In general, the dummy load for testing and 105v rough setting just requires 2.3k resistance ?  Essentially, the psu reads a load roughly that of the mike without the mike in circuit so you set volts and don't fry mike.

Once set, power down, let dissipate, put mike in circuit, pray to Thor, power up, burn in tubes at least 48 hours straight.

That's build finalization process ?

 

[kcatthedog]

Damn,  I was all pumped to dial in my test voltage and I got 10watt not 5 resisters, they equal 2.3k but I am not getting readings anywhere near 105 v.

powered up fine and led is on, no noises and or smoke etc.; certainly throws heat !!

anybody done volt settings with the black zayance ? any pointers appreciated

 

[wave]

Can you post a pic of your build?

Dave

 

[kcatthedog]

Sure from home later,

I have corrected the yellow wire issue to tube socket ( I built mine prior to tath correction being specified) and am tracking power through the  mike, but I get no electricity to the tubes; no light no heat)

I can get 105v on the pcb with dummy load but it jumps to 117-119 with cable (gotham) attached.  I get 119V at red wire in bell but other side of resister (yellow) falls to 9-12v when it should be around 20v ?

I am going to rewire my cable and cat5 is walking me through a diagnostic by email, but I may just open the top of the mike and redo the wiring to the tube sockets.

one step at a time

 

[kcatthedog]

All good, thx to Max, Cat5, Zayance and Stefen, rewired my gotham cable/connectors and fixed a mistake with green wiring to tube socket

105.1v ! and baking my 408a's.

I emailed Max too with the quaity of his builds if my build isn't working I presume its me

Can't wait to try the mike (MK-u47 with blue thiersch cap; find out what that Tony guy is on about !

 

[kcatthedog]

Update:

Thx to Cast5's help, I have been extensively checking my psu and have fully checked wiring (PSU and  cable) and tested for continuity.  I had appeared to have some electrical glitch and I have still not been able to  use my mike. 

To control for all variables,  Cat5 has kindly sent me  a cable he knows is working; should get it next week.

Problem right now, is getting psu dailed back to 105v with dummy 2.3k load.  I had it there last week but after redoing wiring it settled back at 118v with dummy load. It had been there before.

I am using the  230v toroid and have installed the  3rd 1K resister from zayance 's BOM ( so 3  1K plus choke) and removed my piggy backed resisters; oddly the v went up 1v: 117 to 118 ?

Cat5 also recommended I check the variable resister pot indendently from the pcp.  I did and it provides a 1K range and  continously beeps for continuity,  but installed with dummy load, it seems to only scrub off about 1/3 of 1K ?

So, suggestions : - lose the 230v toroid and drop in the  200v, or
                                      - keep the 230v and add more resistance ?
                                      - other ?

My MM was new last year so I do not suspect its readings.

p.s. Really appreciate Cat5's guidance, I have learned a lot about trouble shooting (my mistakes ), he is SO diplomatic!

Thx !

 

[fazeka]

Hi chunger,

Any update on this?

Powering the unit up was uneventful which is good.  No smoke! 

Also surprisingly, my voltages adjusted right to 105V with room to spare on both ends of the range.

The meter supply voltage was low though at about 2.8V with a 35.5K/63.2K divider at the LED supply.  This is where we are drawing power for the 3 wire meter.

Moving the split to 27.4K/68.1K yielded 2.9V which is closer, but not enough for our meter that needs between 3-30V on the supply.

Another change to 18.2K/68.1K produced 4V which is in range.  I like sitting on the lower end of the range because if the PSU is turned on without a load, the open supply will spike up to around 275V pulling the LED and meter supply with it.

The LED is a little bit weak still at 1.7V, so maybe some more tweaking will be in order.


The bigger issue in my opinion is the meter itself which is model YB27_v1.7 (red LEDs) .  The meter is a bit "jumpy" not settling very well.  Contributing to the instability of the reading is the tendency for the passive supply voltage to vary depending on how many LED segments are lit.  The previous prototype which used a YB27_v1.6 (blue LEDs) was better in this regard.  This issue aside, there is an attenuator on the meter and it can be precisely calibrated to 105V which is the point where we require accuracy.  Once the LED segments stabilize, things settle down.  Cleaning the meter with some isopropyl alcohol seemed to help the issue, but I am not sure if it was the cleaning or the new meter.  I purchased quite a few samples and tried changing the LED display on my original meter with yellow one  ;D but that was unsuccessful as the pinouts were different.  The reason I switched to the v1.7 meter is I found it in the 0-300V DC range which suits our application.  But, I would prefer the more numerically stable v1.6 (0-200V) under normal use if I can find it again at quantity.

 

[chunger]

After a lot of testing and swapping meters and resistor values around, Matador decided that powering the meter by the passive and noisr sensitive LED tap in the circuit is not the best way to go so we will utilize one of the 6V taps in the Antek toroid with a simple regulated supply for the meter at 5V.  Alternately, for folks who opt not to use the antek any 6-12v small transformer can be added.

Another prototype and revision may be upcoming and another mk47 psu added to my pile of prototypes 

 

[fazeka]

Thanks for the update chunger!

 

[Inglev]

This kit is gonna be great!

Having just ordered Max' MK-U47 kit, will different psu versions (MK47 and MK-U47) be available? Im guessing the difference is very minor (not needing a pattern switch with the MK-U47), but as a newbie, I feel safer being guided all the way  (like the C12 build!).

Thanks

-Inglev