Shop OBEX P1 Docs P2 Docs Learn Events
PSC and servo power supply — Parallax Forums

PSC and servo power supply

DelHDelH Posts: 5
edited 2006-04-30 00:13 in General Discussion
Can someone please explain to me what is happening in the power supply for the PSC that that the BS2 cannot be powered with the same supply as the servos?·
·
I have a need to have only one power supply and have been playing around with it for a while, with some odd results that I do not understand.
·
First here is the code I am testing with:
·
·
Sdat· CON 0 ‘ßyour pin for PSC
Buff VAR Byte(3) ' temporary variable
·
FindPSC: ' Find and get the version
DEBUG "Finding PSC", CR ' number of the PSC.
SEROUT Sdat, Baud+$8000, [noparse][[/noparse]"!SCVER?",CR]
SERIN Sdat, Baud, 500, FindPSC, [noparse][[/noparse]STR buff\3]
DEBUG "PSC ver: ", buff(0), buff(1), buff(2), CR
·
PAUSE 5000
·
DEBUG "restarting"
GOTO FindPSC
·
·
Basically, the above is just to see if the PSC is responding.·
·
I am running three hitec HS-645MG servos with the PSC and BS2 OEM, and I have tested with two different power supplies with only slightly different results.· I am not running the servo power through the PSC.· Instead, I am only sending it VDD, VSS and P1 from the BS2OEM.· I run one lead from the PSC 1, 2, and 3 along with power and ground from the power supply to the servos
·
The problem is that with all three servos connected to the system, the PSC never responds.· With the above code, my debug window fills up with:
·
Finding PSC
Finding PSC
Finding PSC
·
However, with my 2 amp 6 volt magnetic power supply I have found that if I connect one servo after starting the system, the PSC continues to respond.· But then if I add another servo (without stopping power) the PSC stops responding.· If I add a 3300 micro farad capacitor anywhere in the system, the second servo does not stop the PSC, but adding the third servo again stops it.· Adding another 8200 micro farad cap allows me to also include the third servo.· In all cases, if any servos are connected when I power up the system, the PSC never respond.· It only works with the caps and if I add them to the system one at a time after the system is powered up.· I had thought this was because the caps were drained on startup, but pressing the reset on the PSC doesn’t cause it to start responding.
·
Following the above start up procedure, subsequent tests have shown that there is enough power being supplied to move all three servos at a maximum rate without resetting anything. ·
·
I have had similar results with a second power supply, switched 3.3 Amp 7.5 V but stepped down to 5.93 V with an NTE956.· ·The one difference is that with this setup, adding the third servo still causes the PSC to stop responding.· Later I will test by just adding more caps, but what I need is to do is understand what is going on so that I can set it up to run off one power supply.· I cannot disconnect my servos and reconnect them every time I power down.· Also, these huge caps just don’t seem right to me.· Can anyone please help?

Comments

  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2006-04-25 22:18
    DelH,

    Is there anyway you can post a schematic of what you have? As a sanity check, does it work when you use two seperate supplies?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • DelHDelH Posts: 5
    edited 2006-04-25 23:00
    I just tried two supplies and it does NOT work. The servos do not move at all, however the BS2 and PSC are awake and not resetting. I only have a visio layout of my PCB(actual board). I will need a little while to label what belongs where-i don't have a scanner, so I cannot send a drawing of the schematic.

    FYI-In my earlier test (with a counting loop, I determined that the BS2 was not resetting, it was only the PSC. Thanks. Will post circuit soon.
  • DelHDelH Posts: 5
    edited 2006-04-26 06:27
    Ok I made a mistake when quickly trying to hook up the second power supply. It works perfectly every time when hooked up correctly. So, my setup is exactly the same except instead of the servos getting power from a second power supply I am taking it from from the same line that feeds Vin on the BS2. I took the 4.7K resistors off the PSC earlier when nothing was working-to rule those out as the problem.

    So now, I have been starting and stopping my setup with code that runs all three servos. It works on about 10% of the starts. If it starts, it'll keep running the servos without a problem. Once on, with the PSC responding, I can reprogram the BS2 to have the servos do anything I want. My only problem is getting a reliable startup where the PSC is responding (I have already tested a 2nd PSC with the same results)!

    I looked at my circuit board and have thought better of posting it.....I think it would just confuse the issue. I have quadruple checked, and with the fact that it intermittently works….. On a schematic, mine would look exactly like Parallax documentation with the one difference of using only one power supply. However, there are some functional differences that I now suspect to be the problem:

    1) I do not power the servos through the PSC. Instead, I use a PCB of my own design. I pickup the PSC serial data and bring it back to my board and send it to the servos from there with power from the one supply. As it works intermittently, I don’t see how any of that could be a problem. However, I suspect I made the power lines (on the board) for the servos to small (.040&#8221[noparse];)[/noparse].
    2) Second, there is 18” between the servo controller and my board, the connection is 28 gauge wire. The servos are between 2 and 3 feet away via 26 gauge wire. There are six .100” connectors between the PSC and the servos(serial), 5 between the power supply and servos(supply and ground).

    -OR-

    3) The servos I am using are HiTec HS-645MG, manufactures specs show they draw 9.1 mA at idle, and 450mA running. How does that compare to your re-branded standard servo? When I turn the system on, all the servos make a jump. It could be this sudden draw that is stopping the PSC because the Caps are not charged yet. What can I do about that?

    Outside of making the circuit a wider gauge, do you think there is anything else I should be considering, and should I put the resistors back on (do you recommend them with the PSC), other ideas? Or, is there anything I can do to stop the servos from moving on startup and killing the PSC? Is there a number 4 that should be on my list?
  • Peter VerkaikPeter Verkaik Posts: 3,956
    edited 2006-04-26 12:28
    You can generate a pdf file from any windows program that has a print menu item.
    http://www.pdf995.com/
    This utility installs a pdf printer driver. It does not insert any messages in the pdf file,
    as many other free·utilities do.
    There is·only one (annoying) popup window and a 30 second wait, but this applies to
    the entire document generated. There is no limit on the number of pages.
    You can then attach the pdf file.

    regards peter
    ·
  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2006-04-26 14:22
    DelH said...(trimmed)
    I looked at my circuit board and have thought better of posting it.....I think it would just confuse the issue. I have quadruple checked, and with the fact that it intermittently works….. On a schematic, mine would look exactly like Parallax documentation with the one difference of using only one power supply. However, there are some functional differences that I now suspect to be the problem:
    DelH,

    ·· I would never discount the board as a possibility.· I understand you rechecked your connections and that it works intermittently, but the only way to be 100% sure would be to connect it to a known good board.· For example, if the PSC and BASIC Stamp we plugged into a Board Of Education or PDB with a known good supply and it worked, then you could start narrowing it down, since you could eliminate the PSC and/or Stamp Module from the equation.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Chris Savage
    Parallax Tech Support
    csavage@parallax.com
  • DelHDelH Posts: 5
    edited 2006-04-26 17:17
    I have done the opposite (which is the same thing???), swapped BS2's and PSC onto my board, with the same results.
  • Tom WalkerTom Walker Posts: 509
    edited 2006-04-26 17:56
    So basically what you are saying is that the only thing common in the "not-working-as-desired" arrangement is your board? [noparse]:)[/noparse]

    It could also be that the servos "starting-up" are just demading to much from your single supply (and, thus, potentially "browning out" your other systems). That would explain why a cap allowed you to bump the servo count by one but not two and why separate supplies get better results.

    More info is required.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Truly Understand the Fundamentals and the Path will be so much easier...
  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2006-04-27 00:18
    This was my point exactly...To get the devices to a known good working board.· Not saying your board is definately the problem, but I have had countless numbers of PCBs and board layouts (usually ExpressPCB) sent to me when the designer was positive they followed our schematics exactly or triple-checked everything and in 5 minutes I found an issue.· This isn't always the case, but if you are unable or unwilling the follow proper troubleshooting guidelines I don't know how we'll be able to help you.· You should really try them on a board that is known to work.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Chris Savage
    Parallax Tech Support
    csavage@parallax.com
  • DelHDelH Posts: 5
    edited 2006-04-27 08:10
    Please settle down Chris,

    What I am doing here is trying to understand some more of this stuff; not take the shortest path to a working system regardless if I understand the results or not. I am trying to ask questions, so that I can better understand the direction that you may clearly see is the best. Is there some formal documentation that as you say are "proper trouble shooting guidelines"? Because if there are, then I'd have to apoligize, just direct me to those and I wont have any more questions for this forum.

    I would not mind posting my board design, but it is a board not a schematic! It will not make sense without significant expaination. I am also connecting to three PINGS, 7 LED's via transistors, and the board has 14 resistors and 5 top mount crossover wires and numerous ports for connectors. Also, I have a setup with dip switches that allow me to manually change the order of the LED's----> so without being there or without significant documentation, do you really think you could spot a problem even if it was there?

    So, Tom, yes, it is my setup, but not my board that is the problem. I breadboarded the servo portion of my ciruit with the same results. I have also tested the breadboard with a HUGE 12Amp, 5v supply with the same results (sometimes it works and sometimes it does not). I am a bit confused about this because in a previous posting by Parallax, they said a single power supply was not recommended because power spikes from the servos will cause the BS2 to reset. However, if my system starts, I am able to start and stop and run all the servos at maximum without causing anything to reset. The problem is just on startup. Maybe someone at parallax could just give me an answer like: if stuff is setup the way you explain it cannot fail in that way OR yes, that could happen OR we don't really play with any servos that aren't our own re-branded parallax servos so we don't know.

    In the past I have run up to 8 Hitec HS-311 servos with a simlar one power supply setup (those servos are all in use now, otherwise I would have loved to test those in my current setup). So, I guess I am done posting to this board. Thanks to those of you who did think obout my problem for a little while. I will check in a couple of days if anyone adds anything, but unless there is anything pressing, I am done here.
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2006-04-27 14:40
    DelH,
    You said...
    ...I would not mind posting my board design, but it is a board not a schematic!...
    A picture is worth a thousand words...· Surely you had some intention when laying out your board?· Can't a schematic be derived (reverse engineered) from
    your board that might benefit our understanding of what problems you are experiencing?· It is difficult to understand what you are trying to describe in text,
    and there is too much that can be misinterpreted by a description·alone.· There are many smart people here that are willing and want to help you, but they
    need more information than what you have provided.
    ·
    ·


    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2006-04-27 14:53
    DelH said...(trimmed)
    Please settle down Chris,
    DelH,

    ·· I am quite settled, I was merely trying to be as complete as possible in my observations.· If you perceived this as me ranting or getting upset I was not.· There is no reason for me to get upset about this.· I am content that I am doing my best to help you and that is all I can do.· It is up to you to take the advice offered.· I was just offering suggestions with supporting explanations.· Hopefully in the future these won't be perceived improperly.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Chris Savage
    Parallax Tech Support
    csavage@parallax.com
  • Vern GranerVern Graner Posts: 337
    edited 2006-04-28 22:35
    DelH said...
    What I am doing here is trying to understand some more of this stuff; not take the shortest path to a working system
    A good ambition. I always like to know why something starts working as it helps in future projects and cements my knowledge of what the parts capabilities and limitations are. smile.gif
    DelH said...
    I am also connecting to three PINGS, 7 LED's via transistors, and the board has 14 resistors and 5 top mount crossover wires and numerous ports for connectors.
    Historically, I've found it is very difficult to solve an equation when it contains multiple variables. In order to solve a problem of the type you have outlined, I would start with ONLY the BSII and the PSC. I would then start with the simplest code I could create to verify and excercise the functions of the test system.

    Once I had a working system, I would then add the next "layer" (such as ping sensors or LEDs etc.) and then again rigorously test and verify that each new layer has not introduced failure or created an unexpected interaction.

    Also, as none of us has seen this interesting and seemingly complex design that you described, we can't rule out that some of the features in the design may be interferring, hence the reason for the requests for schematics. Also, if the design is built on a bread board, there are ample opportunities for miswired connections etc. that may contribute to the symptoms. Based on the description you gave of the design, I really would have no way of identifying any particular symptom as belonging to a particular problem. confused.gif

    My advice, reduce the variables and test test test...

    smile.gif

    Vern

    PS: FWIW, there are some *amazingly* talented people on these forums who give away answers they had to purchase though years of experience. In general, polite and gracious interactions result in a quicker trip between problem discovery and a trouble free system. smilewinkgrin.gif I'll close with a quote I think might be relevant:
    Robert A. Heinlein said...
    Moving parts in rubbing contact require lubrication to avoid excessive wear. Honorifics and formal politeness provide lubrication where people rub together. Often the very young, the untraveled, the na
  • Vern GranerVern Graner Posts: 337
    edited 2006-04-28 22:45
    Peter Verkaik said...
    You can generate a pdf file from any windows program that has a print menu item.

    FWIW, In case you're under the impression that there is no *really* free way (i.e. w/o having to shell out $$ to get rid of nag screens etc) to create PDFs for schematic sharing etc, you can do this with an open-source free tool available here:

    http://sourceforge.net/projects/pdfcreator

    and also, Open Office can export directly to PDF as a native function. smile.gif

    Vern

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Vern Graner CNE/CNA/SSE    | "If the network is down, then you're
Senior Systems Engineer    | obviously incompetent so why are we
Texas Information Services | paying you? Of course,if the network
http://www.txis.com        | is up, then we obviously don't need
Austin Office 512 328-8947 | you, so why are we paying you?" ©VLG
  • bulkheadbulkhead Posts: 405
    edited 2006-04-30 00:13
    DelH said...
    It works on about 10% of the starts. If it starts, it'll keep running the servos without a problem. Once on, with the PSC responding, I can reprogram the BS2 to have the servos do anything I want. My only problem is getting a reliable startup where the PSC is responding
    Wow, this is amazing. This is the EXACT problem I've been having for months now (except I use a Javelin stamp).

    Running on 1 battery, I could only get my javelin stamp and PSC to start up correctly a small fraction of the time. Once started however, no amount of servo(s) or load on the servo(s) would cause a reset or the PSC to not respond. I was using a sub-C battery pack, which should be able to supply 30A continuous.

    I also use high torque servos, but have tested with many different ones, still same problem.

    Running on 2 separate batteries, it improves a lot. There is still a 10% chance that, when I switch on my PSC, absolutely NOTHING happens. The servos don't even jerk. Yet other times (maybe another 10%) the servos will go to the PSC's default center position, as if my Javelin weren't even communicating with the PSC.
    DelH said...

    1) I do not power the servos through the PSC. Instead, I use a PCB of my own design. I pickup the PSC serial data and bring it back to my board and send it to the servos from there with power from the one supply. As it works intermittently, I don’t see how any of that could be a problem. However, I suspect I made the power lines (on the board) for the servos to small (.040&#8221[noparse];)[/noparse].
    2) Second, there is 18” between the servo controller and my board, the connection is 28 gauge wire. The servos are between 2 and 3 feet away via 26 gauge wire. There are six .100” connectors between the PSC and the servos(serial), 5 between the power supply and servos(supply and ground).
    I use 16 gauge wire that goes directly from the battery to the PSC's terminals. I get the same problem. My javelin's I/O pin is connected using the 3 wire jumper directly to my PSC. My circuit is about as simple as it gets, and it still has this problem.
    DelH said...

    3) The servos I am using are HiTec HS-645MG, manufactures specs show they draw 9.1 mA at idle, and 450mA running. How does that compare to your re-branded standard servo? When I turn the system on, all the servos make a jump. It could be this sudden draw that is stopping the PSC because the Caps are not charged yet. What can I do about that?
    I have used servos ranging from standard to high torque (comparable to the HS-645MG), and have noticed NO change in success rate of turning on the PSC. Eliminating this variable is easy, just run a single, standard servo. I doubt that current draw is the problem, though. Although I was surprised to see how small the PSC's power lines are-with 16 high torque servos, the current could go past 30amps!

    I'm not really an expert on this, just posting my observations. IMO it is just a bug with the PSC that has yet to be discovered, or maybe just an interferance issue. Either way, the PSC is still a great product. I love the built in ramping feature.

    Post Edited (bulkhead) : 4/30/2006 12:16:40 AM GMT
Sign In or Register to comment.