Image Map Image Map
Page 2 of 4 FirstFirst 1234 LastLast
Results 11 to 20 of 40

Thread: Power Supply grounding

  1. #11
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    29,574
    Blog Entries
    20

    Default

    The 7000 series were from about the mid-1960s. Nice machines. I think I caught sight of one on a SciFi flick the other night--"Gammera the Invincible"--it looked as if it was being used as a newswire service teletype.


    Are you using this manual as your reference?

    It's worthwhile to note that all of the schematics call for the third (ground) terminal of the AC plug to be connected to the frame. So no "hot chassis".

  2. #12
    Join Date
    Feb 2011
    Location
    Salisbury Maryland
    Posts
    763

    Default

    Looked at the schematic in the manual and yes, it’s a three pin plug for AC with the safety ground attached to the frame. The incoming AC is rectified and the resulting DC voltage is used to operate everything but by looking at it you can see at no point is 90 volts source ever tied to ground or the frame and because of this the entire DC control system floats.
    To fix this you would only need to isolate the bridge rectifiers feed by F2 and not need much of a isolation transformer at all. But any attempt to ground one side of the DC control system can be iffy at best and dangerous at worst.
    You are in a sense by doing this tying the neutral back to the ground and would be subject to any differences between the neutral that’s in theory at the same potential to the ground but in practice often different and this will result in cross currents.
    On the FTM series of these machines they install a transformer but still maintain an isolated DC system to allow communication with another unit. By doing this they avoided the issues of noise and cross currents from the power line being introduced.
    It’s also interesting to note that the original design uses no filtering on the DC supply so there is a lot of just random AC junk on that bus along with no bypassing diodes on any of the relays or capacitors on any of the contacts that would result in the DC bus in this thing being almost useless for anything other then what it is. Going to assume that the interface that allowed this to work with any digital device may have relied on fast switching relays to gets around all these shortcomings, but never having seen one of these myself this is all speculation on my part.
    Without building an isolated communications interface I would be real leery of connecting it to any digital devices. It’s not just the “Hot Ground” issue but also the 90 volt control bus in the terminal is somewhat dangerous to any TTL device you will have to contend with the AC noise and short duration transients from switches and relays opening and closing. You may want to try to look at the specifications of the current loop card for your system, this thing is way different from the friendlier ASR-33.
    But that’s just me and what I think, and as my wife would be the first to tell you I am often wrong and will be watching you progress.

  3. #13

    Default

    Quote Originally Posted by Qbus View Post
    Looked at the schematic in the manual and yes, it’s a three pin plug for AC with the safety ground attached to the frame. The incoming AC is rectified and the resulting DC voltage is used to operate everything but by looking at it you can see at no point is 90 volts source ever tied to ground or the frame and because of this the entire DC control system floats.
    To fix this you would only need to isolate the bridge rectifiers feed by F2 and not need much of a isolation transformer at all. But any attempt to ground one side of the DC control system can be iffy at best and dangerous at worst.
    You are in a sense by doing this tying the neutral back to the ground and would be subject to any differences between the neutral that’s in theory at the same potential to the ground but in practice often different and this will result in cross currents.
    On the FTM series of these machines they install a transformer but still maintain an isolated DC system to allow communication with another unit. By doing this they avoided the issues of noise and cross currents from the power line being introduced.
    It’s also interesting to note that the original design uses no filtering on the DC supply so there is a lot of just random AC junk on that bus along with no bypassing diodes on any of the relays or capacitors on any of the contacts that would result in the DC bus in this thing being almost useless for anything other then what it is. Going to assume that the interface that allowed this to work with any digital device may have relied on fast switching relays to gets around all these shortcomings, but never having seen one of these myself this is all speculation on my part.
    Without building an isolated communications interface I would be real leery of connecting it to any digital devices. It’s not just the “Hot Ground” issue but also the 90 volt control bus in the terminal is somewhat dangerous to any TTL device you will have to contend with the AC noise and short duration transients from switches and relays opening and closing. You may want to try to look at the specifications of the current loop card for your system, this thing is way different from the friendlier ASR-33.
    But that’s just me and what I think, and as my wife would be the first to tell you I am often wrong and will be watching you progress.
    I agree. It should be completely opto isolated or relay isolated. I wouldn't even have a capacitor between the two. The schematic shows no ground reference on the DC supply. To connect one would be in error ( a dangerous error at that ).
    Dwight

  4. #14
    Join Date
    Dec 2013
    Location
    Near Milwaukee Wisconsin
    Posts
    1,105

    Default

    I do use opto isolaters between the Friden and the other machines. I have the grounds of each isolated, Yet the 60 cycle noise still gets in.

    Yup, that manual from 1959 and my 1962 schematic are the primary sources of information I use. The manual does not have the model SFD-V in it, but since the manual is 80% mechanical and 20% electrical and most models use similar mechanical schemes, it works for most cases. When electrical items arise, the schematic gives me an idea, but many times my machine is not the same. This manual has the FL (6 bit), FC (5 bit), FG (6 bit), FTM (6 bit) and FPC (6 bit). Fortunately, my schematic is for a 8 bit machine which is much closer to what I have. I'd love to find a manual more specifically for the SFD. I have found some sale literature for the SFD, but nothing more. This is where I found the -V means edge card reader. The name plate doesn't help much.
    Attachment 50424
    Today, I'm working on finding the reason a carriage return code does work. I knocked off a small plastic lever working as a contact actuator. I have to remove some hardware in order to get my hands in far enough to re install it.

    I'll look into removing the neutral from the frame, but since my schematic is not that reliable, it may be real job. Thanks, Mike

  5. #15
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    29,574
    Blog Entries
    20

    Default

    What neutral? According to the schematics in the cited manual, only the ground (third prong) should be attached to the frame.

    On your system, what is the source of the drive for the selector magnet/relay assembly? Do you have a schematic for that?

  6. #16
    Join Date
    Feb 2011
    Location
    Salisbury Maryland
    Posts
    763

    Default

    Looking at the schematic, TC1 is the neutral, TC2 is hot (120) and TC5 and TC6 are the DC output to the relay and switch circuts. But remember that at no point is any DC applied to ground. The relay and switch system is all floating and under normal circumstances has no reference to ground and if you attach to that system with reference to ground that’s where the trouble starts.
    The electrical control system of switches and relays that are being used are more in common with a pre digital Pin Ball machine then any data processing device. A good way to see just how different it is from anything most of us understand these days you can take the DC output on TC5 and TC6 and flip them and the machine will work exactly the same, all they are doing in this automated typewriter tape reader is just simple DC relay and contact control. There is no reason that within the original design they needed to isolate the DC power bus because it was internally isolated in the machine. Unlike teletype machines where you always assumed an external connection to the real world that were all provided isolated loop circuts just for this reason.

    printer.jpg

  7. #17
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    29,574
    Blog Entries
    20

    Default

    That's not what I was asking (I've seen the schematic, obviously). How is he (Mike) driving the relay coils from his CP/M machine? There's got to be a power supply somewhere--and it's probably not in the Flexowriter.

  8. #18
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    2,315

    Default

    Yes, we need the full picture.

    From the schematic posted, I agree that nothing should be connected to the frame/safety earth. Everything is floating. However, nothing is floating in isolation. It is only just floating from the protective earth. You wouldn't see me putting my hands anywhere near any of the wiring - because nothing is isolated from either the mains incoming live or neutral!

    I agree with Chuck in that everything within the Flexowriter should be 'self contained' - so how is it connected to your other machine?

    Just be to be clear - from what I have seen on the schematic, there is no internal smoothing capacitors on the power supply - so (internally to the Flexowriter) the power supply is just full-wave rectified AC.

    Dave

  9. #19
    Join Date
    Dec 2013
    Location
    Near Milwaukee Wisconsin
    Posts
    1,105

    Default

    Here is the output diagram. I only have a hand drawn input circuit, so far.
    FridenOutput.jpg
    The way this works is the translator magnets are selected via port 154 bits 0-8. Then the ZTX957 transistors picks the proper translator magnets. Once this 'code' is picked, the translator clutch is picked via Port 157 bit 0 which types the character. The magnet is picked up with the Friden's 120 VDC supply. I used a triangle ground symbol to show the Friden negative common.

    The input circuits are basically the same. The code selector contacts SC's will light the LED, using the Friden's 120 VDC power. The output transistor will apply either 0 or 5 volts to the Port 155 bits 0-8. This requires a 5 volt supply that is external and isolated from the Friden. I hope to make a better input diagram soon. Mike

    PS, is there a way to post a pdf file here?

  10. #20
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    29,574
    Blog Entries
    20

    Default

    Yeah, I can't read the parts values in the schematic--the resolution is too low.

    Try posting via a link to a third-party hosting service, such as box.com (free).

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •