Hirose nanoZ adaptor for goldplating?

[Kev Kidder]

Hello,

I'd like to switch my headstage/EIB set up to the new low-profile headstage/Hirose EIBs. We currently use the nanoZ with an Omnetics adaptor for gold plating. Does anyone know where I can purchase a 64-ch Hirose adaptor for the nanoZ?

I'm waiting to see if Plexon can make this for me but I'm curious what others have done for goldplating with this new Hirose set up. I'm really excited to test out this low-profile design, the intan headstages gave us lots of issues.    

Best regards,

Kevan Kidder

Mizumori Lab

University of Washington 

[Kev Kidder]

Hi Jon,

I checked with neuralynx, plexon, and white matter. None make a 64-ch hirose adapter for the nanoZ. We paid Plexon ~$260 to make a nanoZ 32-ch omnetics adapter for us in the past. If the cost for your production of this new nanoZ adapter is under say... $600 I would be interested. If more then that, I may try to make it myself. Although I'm nervous about it. This is the last thing I need to figure out before buying the new headstage/EIB set up. 

We would only need 1 ever. Unless it breaks. You mentioned the open-ephys production team may have made some already? 

Thanks,

Kevan

[Kev Kidder]

After consulting my lab we believe we can make the adapter ourselves. From my understanding we will need to:

1) find business to print PCB. 

2) order parts on BOM (3 listed on "nanoz adapter 64 ch" tab)

3) assemble components 

4) test

Any tips for us? 

Thank you for your help Jon. Its very much appreciated! We may reach out if we have further question during assembly and testing.

-Kevan

[Jon Newman]

Hi Kevan,

Here is the design. I will talk with the production site about making these. I think they have already but pandemic, etc, etc.

https://github.com/jonnew/ONIX/tree/main/nanoz-adapter-64

We offer custom design services as well for the record. Get in touch with me next time ;).

- Jon

--
You received this message because you are subscribed to the Google Groups "Open Ephys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-ephys+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-ephys/6013badf-d6cd-451a-b4a1-b8f9d799c608n%40googlegroups.com.

--

Jonathan Newman

Postdoctoral Fellow, MIT

http://www.mit.edu/~jpnewman/

[Jon Newman]

Hi Kevan,

- We can talk via personal email about custom design services pricing

- Honestly, I think the best solution might be for you is to make and assemble the PCB yourself. I can provide some guidance on that if the route you go.

- Jon

If you can get custom design services, assembly, and shipping for 260 dollars then I would go for it. I suspect, however, that the 32-ch omnetics was not from scratch. These things scale with quantity and there is no way plexon turned any profit at all at that price, especially for a non-consumable part. The engineer's salary would have consumed that in an hour or two.

How about this. I will sell you one for 600 dollars, shipping included since I have spare PCBs laying around and I can do the assembly and testing for you. It'll take me a week or two to get the parts. In either case, I will respond with options to the forum question.

--

You received this message because you are subscribed to the Google Groups "Open Ephys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-ephys+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-ephys/80422d78-8dd2-4ca5-9854-859c0061aa22n%40googlegroups.com.

[Jon Newman]

Hi Kevan,

Yes, exactly. For cheap PCBs at incredible prices:

- OSH Park: Small business; hacker centric;  OG of affordable PCB services.

- JLCPCB &  PCBWay: Large scale production facilities in Shenzhen. Truly amazing prices for the quality you get. The base specs are generally not good for high temperature applications because the epoxy they use between layers can't handle it (this is fine for you b/c you can use leaded solder and this won't be used in a high temp environment). However, for the record, these companies can make pretty much whatever PCB quality you want if you are willing to pay them for it. 

As for components, there is one tricky one which is the 70 pin hirose connector. You will need some basic SMD soldering techniques to do this. Look up "drag soldering" on Youtube for lots of tutorials.

- Jon

--

You received this message because you are subscribed to the Google Groups "Open Ephys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-ephys+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-ephys/8efc2a72-b748-49a2-b978-2ce8967c8045n%40googlegroups.com.

[Chris Rodgers]

I mainly agree with what has been said so far, but I hoped to also share my perspective. I've designed and built lots of custom PCBs with OSH Park. It's fun and works well and won't cost too much. But that's not going to be the hard part, I bet.

A few years ago I was in a similar position as Kevan is now, I need an Omnetics to Samtec connector (something like this: https://plexon.com/products/neuronexus-adaptors-hst-32v-64ch/), and was annoyed it cost hundreds of dollars from Plexon or whoever. So I figured I'd design one myself. Here's the PCB I came up with:

https://chris-rodgers.com/cr_uploads/ON1/Image.png

Cost of the PCB: <$10. Cost of my time: Well, let's not think about that part. 

But it turned out designing the PCB wasn't the hard part, it was soldering those tiny legs on the Omnetics connector to it. The problem is that the pins are really close together, AND even worse the plastic connector starts to melt at a pretty low temperature. So while you're soldering the legs, the plastic is melting, and it's a race against time. A hot air reflow station can work but will also melt the connector quickly. Plus each Omentics connector is like $80 or something (i.e., a big chunk of the cost of the entire adapter).

In the old old days (2014?), before you could buy the Open Ephys acquisition board, I assembled one from the raw components using a hot air station. And once again, by far the hardest part was those little Omnetics connectors for the headstage inputs. I still remember how annoying that part was, watching them melt and fall apart.

So my advice is: definitely learn how to design your own PCBs, that's a super useful skill I use all the time for building Arduino / Raspberry Pi shields, or little adapters/interposers so you don't have to have a rat's nest of wires. But try to avoid soldering Omnetics connectors if you can. And if a product already exists that does what you need, think carefully about whether it's a good use of your time to reverse-engineer it.

Just wanted to share my experience, maybe people are better at things that I am and won't struggle with the same things I did, but that's what I have personally learned.

Chris

[Jon Newman]

Hi Chris,

Thank you for this perspective! It's absolutely correct. Grad students and postdocs often do not properly take into account their time when deciding whether or not to do something themselves. It can be a bit shocking when you get a quote for a one-off design as simple as this, but more often than not, it's because the engineer knows something you don't.

That said, one of the best features of this EIB design is that it does *not* use omnetics!!!! Thank goodness. They are horrible parts as you have discovered. This company, which neuroscientists have come to rely on

1. Make all of their real money off of the US military industrial complex

2. Have terrible QC for smd components -- pins are made from what appears to be soft cheese and are often bent all over the place right out of the box.

3. Use very low-melt point plastics for some reason.

4. Cost 50-100 USD for a _single connector_ with a _1 month_ lead time. 

I couldn't think of a worse mix. The hirose connector on this EIB is going to be extremely easy to deal with despite its tiny pin pitch because it isn't designed by an insane person.

- Jon

To view this discussion on the web visit https://groups.google.com/d/msgid/open-ephys/cb941f80-f9c9-40c3-a4c6-19b29ae691e4n%40googlegroups.com.

[Kevan Kidder]

Hi Jon,

I'd like to update you on how it has been going with making the Nanoz -64ch adaptor. We have 1 minor issue with part ordering I wanted to correct for you in the bill of materials. Also we have a critical issue Im curious if you know about and if there is a way around it.

1) (minor) We got PCB boards made by OSH park. Bought all necessary pieces and found that I had ordered one wrong part as the link in the Bill of materials is for a slightly different part then is listed/needed.  The current link is for Part# MOLC-120-01-L-Q . However what is listed as the part number and is the correct part is  MOLC-110-11-S-Q. Here is the correct link for this part # https://www.samtec.com/products/molc-110-11-s-q  

2) (critical) We waited a month for this correct part to get in and once we did we are finding that 1 of the quad row terminal strips fits nicely in its respective vias, however the other quad strip is slightly misaligned and it's connectors/pins sit right in the middle of the PCB vias. So for us it appears that one full group of the vias are about .5mm too far anterior. Below are some images. Im guessing its not ok to bend these silver pins to make it fit, correct? We also have the "nanoZ model: 1.3" in case that could be affecting it. Have you experienced this issue or have any tips for this?

Front/anterior image

Best,

Kevan Kidder

Mizumori Lab

[Jon Newman]

Hi Kevan,

This text, verbatim, is a good github issue. Please feel free to put it up on the repo if you want. See below.

- Jon

On Thu, Jul 29, 2021 at 4:39 PM Kevan Kidder <kevki...@gmail.com> wrote:

Hi Jon,

I'd like to update you on how it has been going with making the Nanoz -64ch adaptor. We have 1 minor issue with part ordering I wanted to correct for you in the bill of materials. Also we have a critical issue Im curious if you know about and if there is a way around it.

1) (minor) We got PCB boards made by OSH park. Bought all necessary pieces and found that I had ordered one wrong part as the link in the Bill of materials is for a slightly different part then is listed/needed.  The current link is for Part# MOLC-120-01-L-Q . However what is listed as the part number and is the correct part is  MOLC-110-11-S-Q. Here is the correct link for this part # https://www.samtec.com/products/molc-110-11-s-q  

MOLC-110-11-S-Q

 

The bill of materials has the correct part number. 

 

2) (critical) We waited a month for this correct part to get in

Samtec will send you free samples in 1 day, at least in the US, no questions asked. I basically never buy anything from them. Sorry Samtec if you are listening. 

 

and once we did we are finding that 1 of the quad row terminal strips fits nicely in its respective vias, however the other quad strip is slightly misaligned and it's connectors/pins sit right in the middle of the PCB vias.

 

I appreciate the pictures, but I'm still not exactly sure what is happening. On our end, You can look at the design files and see that the pitch is the same between the two connectors, so there is no way for this to happen barring some extremely strange scaling issue at the manufacturing level. If the pins are not lining up with the socket on the Nano-Z, it's possible that there was an interface-breaking revision but that would be a terrible idea from a business perspective. We use this adapter on a daily basis in our lab.

 

To view this discussion on the web visit https://groups.google.com/d/msgid/open-ephys/e8a1b345-b07d-4b7e-a4fb-f35a0ddacee8n%40googlegroups.com.

[Jon Newman]

 interface-breaking revision ->  interface-breaking revision _of the nano-z itself_

[Jesse Miles]

Hey Jon,

Sorry for the confusion, I'll try to clarify. The issue doesn't seem to be with the pitch on the PCB, but with the spacing between where the two connectors are supposed to attach. It looks like it is off by ~0.5 mm. The Samtec connectors themselves will individually fit into the PCB you designed just fine, but the PCB with the connectors attached will not connect to the nano-z because there is too much space between them. Likewise, in the images Kevan sent, we tried attaching the connectors to the nano-z first and then attaching the PCB to the connectors, which gives the mismatch Kevan showed in his pictures - everything is perfectly aligned on the right connector, but the left connector is shifted slightly right of the rightmost vias on the PCB (and the shift is constant for every connector lead). 

I'll create an issue on your github repo as well if that's preferred, but can you let us know what version of the nano-z you have just in case it is, in fact, a weird interface-breaking revision on the nano-z's end?

Thanks!

Jesse Miles

Gire and Mizumori labs

[Jon Newman]

Hi Jesse,

Like I said, we use this all the time so I just dont really understand what is happening outside of the nano-z itself changing the spacing. That would be very strange because it would make all the adapters people have no longer work if they had to buy another one. I will double check the revision number on the one we have in lab to make sure I didn't do something very strange, but its seems crazy that I would have a working adapter then go in and change the spacing randomly. 

- Jon

To view this discussion on the web visit https://groups.google.com/d/msgid/open-ephys/7fe73f24-8636-48ab-b971-df2e43164a43n%40googlegroups.com.

[Chris Rodgers]

Hi all

Resurrecting this old thread because I am now also in the market for an adaptor to connect the low-profile OE EIB with Hirose 64ch connector to a nanoZ for impedance check and goldplating. 

Wondering if getting this design (https://github.com/jonnew/ONIX/tree/main/nanoz-adapter-64) printed by a place like OSH Park is still the recommended path forward? Or if Open Ephys now sells the completed product directly?

I have been emailing Open Ephys sales already with the same question, so I'll post here if they know the answer, but it occurred to me that the google forum might be a better route. 

Thanks!

Chris

[Kevan Kidder]

Hi Chris,

We at first had the PCB board made by OSH park. For us, there were a few pieces that were difficult to put together/we didn't feel lined up well. We ended up buying a fully assembled one from open ephys which works great. 

Best,

Kevan Kidder

Ph.D. Candidate

Mizumori Laboratory 

University of Washington 

You received this message because you are subscribed to a topic in the Google Groups "Open Ephys" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/open-ephys/xTc9H_jmoxg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to open-ephys+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-ephys/1b6d5585-0e1a-4681-b41b-f90156d1e87fn%40googlegroups.com.

[Jon Newman]

Hi Everyone,

The adapter is available here if you want to buy it assembled: https://open-ephys.org/acquisition-system/nanoz-64-adapter

We have plans to make our website easier to navigate. Oftentimes even I cannot easily find things in the store.

- Jon

To view this discussion on the web visit https://groups.google.com/d/msgid/open-ephys/CADEnTyazE4xTmoo7%3DBjH578hxReMN6iQsfJ%3DVm8KjGNpZMTdcw%40mail.gmail.com.

[Chris Rodgers]

Oh, it was right there all along. Thanks Jon!