A10 Rack modification (for ADX5D)

[Wyatt Tuzo]

So... I started my transition into the Axient Digital line earlier last year. For months I had been sitting on a number of transmitters, in the hopes of a receiver solution that would suit the layout of my cart. I live and work in New York, and thus I've always valued having as mobile, and lightweight a cart as I can get away with, and the full-sized Shure rack receivers just weren't going to work with my setup. 

Further considerations of mine were: The hope for DANTE integration, and a relatively high channel density (higher than Shure's AD4Q, at least). 

 

When the ADX5D was announced, I figured all of my problems would be solved by pairing these with the Audio Ltd (now Sound Devices) A10 Rack. Unfortunately, the Shure receivers were just a bit too physically broad for the A10's SMA antenna spacing. Inspired by a recent Facebook discussion on the Axient Users Group, I decided to look into modding an A10 Rack to accommodate the ADX5D.

 

I thought I'd share my process in the event anyone is curious... I'm going to spread this out over a couple of separate posts below.

 

Please know that this modification is not for the faint of heart. I feel pretty confident working with electronics, and I was still triple and quadruple checking everything as I went. While this may present as a how-to, I am not looking to encourage anyone to try this for themselves. I accept no responsibility for your choice (should you attempt this), and you will most certainly void your warranty.

 

Here goes...

[Wyatt Tuzo]

Step 1: The Plan

 

 

 

 

You can see here how the SMA's prevent the receiver from mounting properly:

 

 

Again, inspired by a discussion on FB, there seemed to be two possible ways I could go. I could A: Install DP25 spacers and custom fabricate receiver spacers that would make up for the extra distance to the chassis, or B: Replace the SMA jacks altogether.

I thought I should have a look at option A first, as it would be the least destructive.

 

After Some Dremeling, I was able to get the DB25 spacer to mate properly with the receiver backplate:

 

 

 

While this would certainly have been the easier course, there were a few things that didn't quite sit right with me. Firstly, the receivers ended up standing a bit too proud of the chassis for my taste, and second, the DB25 spacers would have no way of securing to either the chassis or the Rx unit. As a part of the plan here is to be able to pull receivers to go into a bag when needed, I felt this was a potential problem.

 

Let's go with Option B

 

 

 

Step 2: Proof of concept

 

I decided to replace the SMA connectors with lengths of RG-174 that would home-run to the PCB. I added grommets for safety and fit:

 

 

 

With everything looking good so far, I decided to take a quick measurement of the RF from the antenna input to the SMA connectors for comparison later (**Note the Blue line and value, aprox 2 and 3 dB of gain, respectively, within the unit at 522Mhz. There is no significance to the freq I was metering, it just happens to be where the curser was at the time):

 

That was supposed to say "Ant. B", but my phone autocorrected and I just noticed. I'm going to live with it.

[Wyatt Tuzo]

Step 3: Pull everything out

 

Carefully remove the DANTE board, then remove main PCB screws and DANTE card stand-offs:

 

 

Remove receiver slots by gently pulling PCB away, then removing all screws:

 

 

 

Look at those tiny 0402 sized SMD components directly next to the work area! My guess is that these are either for filtering or buffering.

[Wyatt Tuzo]

Step 4: Preparing the PCB and faceplate:

 

As the grommets don't sit completely flush with the faceplate, and this will need to mount directly against the chassis front, I'm unfortunately going to have to drill out the chassis to accommodate the outer diameter of the grommets:

 

The grommets measure 5/16... I'm going to drill the chassis holes to 11/32. I take extra care de-burring these holes after, and blowing out any potential stray metal filings... that would be the last thing I'd want bouncing around in here once I'm finished:

 

I don't have photos of the next step (removing the SMA connectors) because I decided to take the board to a specialist for that. I was too nervous given the proximity of the aforementioned circuit... Money well spent. Here is a pic of me cleaning the residual flux off the PCB afterward:

 

 

Note that even the pro's pulled off a few solder pads in the process. Fortunately, they were just ground pads and I had enough to spare:

 

 

 

[Wyatt Tuzo]

Step 5: Making all connections

 

I decided to cover the neighboring circuit with Kapton for some added insulation prior to adding my RF leads:

 

After meticulously making my RF leads (using plenty of heat shrink for further insulation), I decided to add more Kapton tape, both over the connections, and down the PCB arms a bit, to act as a strain relief:

 

Next, I re-installed the receiver slots so I could secure the faceplate to the front of the chassis (the receiver slots attach via screws under the faceplate, as well as inside the unit). Note: I used blue thread lock on EVERYTHING while reassembling. This is going to bounce around on a cart, on a truck, hopefully for a few years to come. I didn't want anything shaking loose:

 

Next, I fed the other end of the RF leads through the grommets, and terminated with RA SMA connectors:

 

[Wyatt Tuzo]

Step 6: Finishing/testing

 

Time to measure again! Two things to note: I realize that the curser here is at 520 Mhz, and not 522 as above, but this is good enough to give me a picture and tell me that everything is working as expected. Also, I think the the greater difference in Ant. B from the before may point to something being off in my before measurement . I feel like it's more likely that the two antennas legs were not measuring off by a full dB from the factory, and that the after is probably more indicative of what to expect from the unit. So, I'm happy with this:

 

 

 Note the Excess cable that I left inside the unit. I did this for a few reasons: it allowed me to work more comfortably, I didn't have to be quite as precious with my cable measurements, the slack takes any potential strain off the connection to the PCB, and if I ever have a problematic connection, I have extra cable that I can just pull through and re-terminate:

 

Viola!:

 

I haven't had a chance to pass audio yet, as I left all of my transmitters at my office, but a receiver tone-up is working as it should on all DANTE outputs.

 

I think this means I can move forward with finalizing my transition now (once ADX5D units become available again). My plan is to go through this process for a second A10, giving me 16 channels of Axient in 2u!

 

One more thing to mention: I'm going to have to mill a notch out of the DB25 riser for A10 position 1 in order to accommodate the power switch. Fortunately, the receiver will sit just above the switch bezel (as I'd like to keep the bezel to prevent accidental bumping)

 

Whew!

[Matt Radlauer]

This is FANTASTIC!! I might be hitting you up soon..

[vin]

wonderful... 

 

Apropos getting those SMA connectors off - imho the best way is to use a clipper and cut off the legs soldered to the board. and then desoldering those pieces of metal off the board one by one becomes very easy. 

 

cheers, 

-vin

 

[Wyatt Tuzo]

On 1/24/2022 at 1:13 AM, Vinod Subramanian said:

wonderful... 

 

Apropos getting those SMA connectors off - imho the best way is to use a clipper and cut off the legs soldered to the board. and then desoldering those pieces of metal off the board one by one becomes very easy. 

 

cheers, 

-vin

 

Good angle, Vin. Maybe I'll give that a shot with my next one. 

One thing that I didn't mention in my original post that is worth noting, is that the SMA connectors are soldered to both the top and bottom of the PCB, for a total of 5 connection points. This would leave hot air desoldering, your method, or a product called chip-quick (which reduces the melting point of solder in order to give you more work time) as options. The hot air is tricky without a microscope on account of the circuit near the SMA's, leaving the last two as the most practical to execute at home

[vin]

Hello Wyatt

it is exactly because of these multiple connection points (one is signal and the others are all ground/chassis legs) that it is cutting off the legs is the best way. Even chip-quik would be complicated because of dual side solder points and also the distance between the solder point makes chip-quik not a real solution here...  

 

cheers, 

-vin

[Martin Lonek]

On 1/23/2022 at 3:08 AM, Wyatt Tuzo said:

Dear Wyatt, thanks for a great DIY work ! I´m the the Sennheiser EK 6042 (hybrid digital and Analogue receiver in one) user. It has the same issue ! So your solution can be used for this one, too ! Thank you for creating a manual for us, too !</br></br><a href="www.conference.cz/pic/IMG_0208.JPG">see picture here</a>

 

[Wyatt Tuzo]

18 hours ago, Martin Lonek said:

 

 

You're very welcome, Martin!

One thing I've been hoping to follow up on... after the conversation with Vin above, I decided to try snipping the SMA's off of my second mod and the traces flew right off. This is really a job for a focused hot-air rework station. The ground pads are extremely weak on this board. Fair warning.

 

In other news, I was thinking of machining notch for the db25 risers for the 1st position receivers, but decided to 3d print some instead. I'd be happy to share the file, if anyone needs, but know that I am no expert in 3d rendering 

[vin]

On 4/27/2022 at 8:06 AM, Wyatt Tuzo said:

 

One thing I've been hoping to follow up on... after the conversation with Vin above, I decided to try snipping the SMA's off of my second mod and the traces flew right off. This is really a job for a focused hot-air rework station. The ground pads are extremely weak on this board. Fair warning.

 

So sorry to hear this! 

Then the solution is ChipQuik!  

[Wyatt Tuzo]

Long delay here, but I finally got some downtime and was able to integrate everything. All is working well, and as hoped.

 

Racks are communicating over Dante, with Showlink intact.

 

[Dejan Ceko]

7 hours ago, Wyatt Tuzo said:

Long delay here, but I finally got some downtime and was able to integrate everything. All is working well, and as hoped.

 

Racks are communicating over Dante, with Showlink intact.

 

 

 

Astonishing work !

[Patrick Farrell]

Really nice work Wyatt. Half the space of the AD4Q. Only downside is you need the tablet/computer for easy settings management compared to the AD4Q.

[Wyatt Tuzo]

Thanks Patrick. I remember when you and I first opened one of these up to assess a plan of attack for this. 
 

While you’re right on the drawback of not quite having the control in one convenient place, my tablet will generally be positioned in front of the Rx rack, giving me even more complete control and feedback. That said, there are more potential points of failure. 
 

In the pros column, I get: modularity, a more shallow and lighter rack, and greater channel density/rack

 

Cons: more potential points of failure, and less integrated networking capabilities. 
 

An example on the last point… in order to use showlink to control all of this, I have a wifi access point and separate PoE injector built into my cart. One nice aspect of this, is that either myself or my Utility can have access to settings through WWB and/or Channels on our individual devices. A downside, is that I have to keep a 610 in range of my cart or I lose showlink connection to the Rx. There is a slim possibility that this will drive me to get an additional 610 to position close to set for Tx control in the future. This will iron itself out with time. 

[Patrick Farrell]

Yeah kind of a bummer you can't just remote the 610 when needed and instead would need a second unit. However, my experience was we rarely needed showlink control anyways. If we needed new frequencies, I'd just do it at the beginning of the day with the transmitters still at the cart. Only other use was to sleep/unsleep the transmitters for lunch or a long down time and we'd just have talent walk past the cart. But even with them close sometimes that could be troublesome and it wouldn't pick up the showlink. Separate issue I told the guys at Shure about but I'm trying to make the argument that a second 610 would only be useful in a very small use case.

[Jobo]

Wyatt, I’m interested to hear how your setup is working now several months later since there has been no new Dante rack/slot options added to market by the major players.  Any additional mods or discoveries now that it’s been field tested assuminggly for a good stretch? 

 

[Wyatt Tuzo]

No news, really. They’ve been working reliably for a few years now. I have no reason to suspect that they won’t continue to work reliably. 

My rx are not screwed down into the slots (this allows me to pull them easily if I need to move to a bag), and there has been plenty of friction to support them staying in place over many, many bumpy NYC truck moves. 

 

So, In short… all good.