PDA

View Full Version : Plastic cover for the HB-25 PCB.



Bamse
07-10-2011, 11:39 PM
Howdy...

I just managed to burn out one of my HB-25s.
With all the driving I was doing, one of wires connecting the motor to the HB-25 came loose and dragged over the PCB.

Needless to say, there were some magic smoke produced and now the fan does not come on and the motor is stuck in full throttle forward...

Is it possible to make some kind of plastic cover for the HB-25 PCB to protect it from loose wires ?
Have anyone else had this problem ?

I guess my best bet for now is to start use thread lock to keep the wires from shaking loose...

Tom C
07-11-2011, 12:32 AM
Howdy...

I just managed to burn out one of my HB-25s.
With all the driving I was doing, one of wires connecting the motor to the HB-25 came loose and dragged over the PCB.

Needless to say, there were some magic smoke produced and now the fan does not come on and the motor is stuck in full throttle forward...

Is it possible to make some kind of plastic cover for the HB-25 PCB to protect it from loose wires ?
Have anyone else had this problem ?

I guess my best bet for now is to start use thread lock to keep the wires from shaking loose...

Bamse,
In Aerospace we have planned periodic maintenance which means checking that all of the screws are tight and making sure that the nuts have not backed off their threads on a regular basis.

I do the same with my model planes. I always check the propeller collet and the servo links to make sure that nothing has come loose that could cause loss of control of the plane.

Just a thought.

Regards,
TCIII

Bamse
07-11-2011, 03:53 PM
Yes, I've learned my lesson, I'll be checking those screws in the future.
However that does not help if the robot bumps into something and something falls on top of the PCB.

Would it be safe to spread some epoxy over the PCB ?
That would help but would it affect functionality ?

RobotWorkshop
07-12-2011, 05:18 PM
As long as the screws are tight they shouldn't come loose. If you are concerned about it you could always add a small drop of locktite on the screw threads. It will make it difficult to remove later but will keep them from coming loose.

If you are concerned about stuff falling on the HB-25's then you can make your own covers. Just make sure they are vented so the fans can do their work. I've used some of the small plastic light switch boxes at Home Depot (or Lowes) and even some small plastic food containers.

Coating the boards would be a last resort and I would avoid that if you can.

Robert

Ken Gracey
07-13-2011, 11:14 PM
Heya Bamse,

Let's get that HB-25 replaced ASAP.

Drop me your shipping address via e-mail (no PM, don't get to them often enough) and we'll fix this problem right away. I'd rather replace it than make a plastic case right now because all the machines (and more importantly, the machinists) are running full-throttle and if I ask them for anything else I'll face a full mutiny or even a scarring by cat40 tool holder loaded with a Kennametal fly cutter.

Get some threadlock, too. I've been using that nicely on my Madeusa robots.

Ken Gracey

Bamse
07-21-2011, 05:20 PM
Hi Ken,

I talked to support last week and my HB-25 arrived yesterday.
Thanks for replacing it.

I got everything mounted on the robot and it works great, threadlock is applied and I'm all set for doing some programming this weekend...

Thanks again,
Fredrik

John R.
07-22-2011, 02:53 PM
Personally, I disagree with thread lock on electrical connections. In addition to potential problems with repairs/replacement/re-design, etc., the thread locker is an insulator. Also, given potential thermal cycles, and distortion of connections, you will still need to tighten connections from time to time. If the threads are "locked", this may not be easy to do.

The core problem here is really the power connections on the HB-25. They are possibly the worse connections I've ever seen for a 25 Amp (35 surge) connection. They only way to get an even somewhat reliable connection is to find some type of ring connector (i.e. crip on or solder on) with a small enough hole to fit the screw. Most of the small hole connectors I've found readily available have too big a hole, and don't proberly fit the terminal.

Trying to use a spade connector is not a good fit, as the post is too small in diameter to get good mechanical clamping. Trying to directly clamp the wire between the screw and post is asking for disaster.

This connection needs to be upgraded to provide more area for the connection and clamping a terminal. The connection may be sized adequately for electrical capacity, but there appears to have been little or no concern/thought to the mechanical part of the connection.

PJ Allen
07-22-2011, 06:33 PM
John R.,
No need to worry, because if anything happens to it then you get a free replacement.

RobotWorkshop
07-23-2011, 09:53 AM
Personally, I disagree with thread lock on electrical connections. In addition to potential problems with repairs/replacement/re-design, etc., the thread locker is an insulator. Also, given potential thermal cycles, and distortion of connections, you will still need to tighten connections from time to time. If the threads are "locked", this may not be easy to do.

The core problem here is really the power connections on the HB-25. They are possibly the worse connections I've ever seen for a 25 Amp (35 surge) connection. They only way to get an even somewhat reliable connection is to find some type of ring connector (i.e. crip on or solder on) with a small enough hole to fit the screw. Most of the small hole connectors I've found readily available have too big a hole, and don't proberly fit the terminal.

This connection needs to be upgraded to provide more area for the connection and clamping a terminal. The connection may be sized adequately for electrical capacity, but there appears to have been little or no concern/thought to the mechanical part of the connection.

I haven't had any issues with the connections on the HB-25's and had no trouble finding ring terminals that fit perfectly. The local terminal supply company and auto parts store had them sitting on the shelf. Also Mouser and Digikey can source them easily. I agree that the proper connector to use is a crimp on Ring terminal. That way even if the screw were happen to losen up the wire won't fall off.

I don't use locktight on the threads but as long as you get them tight you shouldn't see them losen up under normal use. Even with some thermal expansion that shouldn't be extreme enough to make them loosen up anytime soon. Unless you are going to use these on a battlebot or some other robot subject to extreme abuse you probably don't need locktight on the threads. If you do use it I would use it sparingly and just have a drop on the threads. Even though it may act as an insulator on the threads themselves, there should still be plenty of contact from the surface of the ring terminal to the metal support it is screwed to.

While having a cover for the board may be a nice feature I really don't think it is required.

The only thing I would like to see in regards to the HB-25's is an update to the docs. There needs to be a note that the mounting screws and heat sink are connected to the ground terminal. It wasn't mentioned in the docs. I found the trace on the board and ended up discovering that myself. The fact that they are grounded may affect how people mount them.

Robert

PJ Allen
07-23-2011, 03:46 PM
Locking Fork Terminals (http://www.grainger.com/Grainger/3M-Locking-Fork-Terminal-1A084?Pid=search) along with a lock or star washer and you have the equivalent of suspenders and a belt!

John R.
07-24-2011, 12:47 AM
The problem with the locking forks is that I have not seen them for a #4 terminal (as used on the HB-25), only down to a #6.

Same problem with Ring terminals, finding #4 rings "locally" is not always going to be possible.

At the risk of being pushy - I still maintain that the power terminals on the HB-25 are lower than "sub-standard" in terms of design, functionality and useability.

RobotWorkshop
07-25-2011, 02:50 PM
[QUOTE=John R.;1020680]At the risk of being pushy - I still maintain that the power terminals on the HB-25 are lower than "sub-standard" in terms of design, functionality and useability./QUOTE]

Well, there's nothing wrong with suggesting they could be improved if there is something you don't like. What kind of terminal would you suggest they use instead? The current terminals have worked ok for the application that I've used them on but everyone has different needs and perhaps your application has more demanding needs.

John R.
07-25-2011, 06:03 PM
Well, there's nothing wrong with suggesting they could be improved if there is something you don't like. What kind of terminal would you suggest they use instead? The current terminals have worked ok for the application that I've used them on but everyone has different needs and perhaps your application has more demanding needs.

How about we start with why a #4 terminal? How many screw type electrical connections you ever seen with a #4 screw. Sorry, while available via Digikey and other online retailers, this size is not commonly available outside metro areas, at least in WisCOWnsin.

Having said #4 screw terminal on top of a standoff that is smaller in OD than a #4 Ring Terminal is not appropriate. The standoff is small enough that if you use something like a #6 ring terminal, or even a #4 spade terminal, you are not likely to get a good clamping surface.

So, what I feel should be there would be #6 terminal screws, either in a propper terminal block or on an appropriate circuit board pad if, or on top of a standoff at least as large as an appropriate sized ring terminal. Even better, how about a real live terminal block, such as that found on the Prop Pro Development board (and other products)?

While when used with the correct terminal, the current design "works", this does not mean that the design is appropriate. "Working OK", is not the same as "appropriately designed". Let's also remember these are supposed to be targeted at a "hobby" audiance, not a full fledged EE. I don't think it's unreasonable to expect a user would attempt to wrap a "twisted" bare wire in the terminals. I'm not saying that this would be an appropriate way of doing things, but it would be appropriate design criteria to anticipate that a user of this device would attempt to do it.

If the discussion is "what will work" or "what can be made to work", these are all mute points. If the discussion is how to bring the HB-25 up to the usual Parallax standards in terms of reliability and ease of use, I believe these are all valid points.

Other points that should also be made with respect to "usual Parallax standards" include the common ground with the mounting holes, the mounting method in general (requirement to either have access to the back side of the mounting, or the adapter mounting plate), the fact that if you don't have a signal there before power up of the HB-25 it will be "dead" (as opposed to assume "off" until a signal is received). Of a lesser concern, but still something I'd change would be to have some type of a stand off from the face of the fan so that you don't risk overheating if you mount the fan against another surface.

There seems to be a newer breed of engineers involved in some designs at Parallax that do not grasp the value Parallax has build in making things "bullet proof" and easy to use, and hard to mis-use. The HB-25 being a prime example. A device, that when applied properly works great. When not all the little twists and turns are applied properly, it can work not so good. I believe that if this product had been designed by the "old school" thought process, it would work equally well when somewhat misapplied.

PJ Allen
07-25-2011, 08:06 PM
Maybe they'll incorporate #6 studs in the next (?) rev. or screw-down terminals or whatever's contemplated.
Maybe Parallax should source the ideal/recommended terminals?

In the meantime, in between time,...
here's a 20th-century solution --
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=7311K-ND

If you need a large-pattern flat washer to augment the mounting surface/s then go ahead.

Bamse
07-27-2011, 05:14 AM
I like those Terminal lugs, might have to order few...

No problems with ring terminals and thread lock, the robot is working fine now.