[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [N8VEM-S100:5883] Re: S-100 8-slot Backplane Version 2 Questions



Hi Bob,

I'll be building one of these boards soon.  Thanks for all your notes regarding your build.  I'll likely go with discrete resistors.  I haven't looked at any plans, I just thought this was a plain 8 slot board.  Are all these extra parts to terminate the S-100 bus?  F1 sounds like a fuse?  There are fuses that look like 1/4W resistors and some even smaller, SMT.

Cheers,
Josh



Date: Tue, 23 Dec 2014 05:47:50 -0800
From: bbel...@gmail.com
To: n8vem...@googlegroups.com
Subject: [N8VEM-S100:5883] Re: S-100 8-slot Backplane Version 2 Questions

Good to hear that you're going down the same road that I am.  I'm hoping to hear from the board designers (Tood Goodman thinks it was Andrew Lynch and/or John Monohan) regarding the issues I raised.  Until then, I have found some solutions:

First, the 8-pin DIP socket could be done using a wire-wrap type socket as you suggest; I have not tried but I fear the pins may be too large (and square) to fit through the holes in the board.  I have planned to use (ordering soon) a Machine Tooled solder tail socket I found at Jameco.  The manufacturer is unknown (JamecoValuePro), but machine tool sockets, in my opinion, are always very good but expensive.  The part number is 51626 and the cost is $0.49 each.  This part has a through-hole lead length of 0.177" and exceeds the board's 0.14" thickness rather nicely.

Next, the 2 and 3 pin jumper headers: I found that the pins are not totally rigid in the plastic.  If you try hard enough, they will slide. I am planning to carefully slide the plastic just a bit so I have enough of the pins protruding through the hole, then solder the pins.  At this point, the top side pin length will be just a little too short to make reliable contact with jumper shunts, so, since the plastic is optional and actually no longer needed with the pins now soldered, I'll use my dikes to cut/break the plastic away from the pins, leaving a satisfactory amount of material for the jumper shunts.

Resistor networks: Every R-net I looked up has a lead length of 0.135 +/- 0.005.  This won't work; I tried one I had here in my junk box, albeit the wrong value, and the pins definitely do not protrude.  Todd Goodman suggested a Military version with the part number M83401/24.  Looking this up on Mouser shows this to be a DIP network, so not useable.  However, I dug a little deeper and discovered the M83401 is like a generic Mil spec number.  Then I found (on Mouser) the exact part Todd was talking about with 0.195 lead length.  But it is non-stocked with a 12 week lead time and costs $2.14 each.  The complete Vishay part number is MSP10A01270RGDA.  To leave no stone unturned, I search Ebay.  No dice.  So, perhaps if future interest warrants, and the board thickness remains at 0.14 (I rather like the rigidity), we might want to consider a group purchase of this part.  Until then I  think my only recourse will be to "build" my own R-nets from discreet resistors.

I also checked out the pushbutton switch for lead length.  I think the E-Switch TL1105-A from Jameco will fit.  The lead lengths are 0.138, but there are "retention" bends in the leads.  I'm hoping that by bending the leads straight, there will be just enough to solder.  At $0.29, I'll not be too sad if it doesn't, but then I'll be back to looking again.  Fortunately, I can use an external switch on P12 as an alternate.

I raised a question regarding the orientation of one of the transistors.  This is based on a note in the Wiki: "Fix Q1 transistor orientation. Involves a 180 degree rotation."  I don't know if this was an actual issue, or with a particular board version, or if it was fixed in the Version 2 I have.  I'll be sure to examine this closely before assembly so the part goes in properly.  I will post another note when I make a determination.

Since I have not heard from the board designers, and there are conflicting photos of the board (one showing heatsinks, one not), and I really don't feel like analyzing the circuit to determine need, I am just going to buy heatsinks for the TO-220 devices and put them on.  At $0.29 each from Jameco (part number 577102B00000G), it's not worth the time.

Please note that the LM4250 Op-Amp is no longer carried by Jameco.  Unicorn electronics has it for $0.49, but they have a $25 minimum.  Mouser also has it for $1.20 with lowest possible freight charge of $4.99 with a 8 day delivery.

Some other random notes:
I am not populating P2, P3, P4 or P5 - unneeded in my application, but will have lead length issues like the 2 and 3 pin headers described above.
I still don't understand the functions of F1, F2 and F3.  I'll be populating them with 20 or 22 gauge wire.
Since I don't have the capability to terminate wires into the housings needed to mate with the header designated for P1, I found a great replacement - terminal blocks.  I had some Jameco 152354 left over from an earlier project and found that they fit perfect.  Now my wiring from the power supply to the board is simply discreet wire, stripped and tinned at both ends.  (My power supply has screw-terminals.)  Unfortunately, Jameco does not seem to be stocking this part any longer, but there are many similar items in the catalog, just be sure to get something with a 0.2" (0.197") pin spacing.  The ones I have are sold in 2 and 3 position, and they snap together to make longer length.  I am using 2 three-position parts.

I hope this helps your efforts (and anyone else building this board).  I'll post any updates as they come up.

Bob Bell








On Friday, December 19, 2014 4:56:35 PM UTC-5, hl...@aol.com wrote:
Hi Bob,

I am also in the process of building a 8-slot  backplane version 2. I also ran into issues with the thick board. I ended up using a wire-wrap 8-pin type socket so that the leads would clear the board. I also had problems with the 270 resistor networks. The leads barely cleared. What did you do?  Unfortunately I didn't check the schematic before I started so if the LED is too dim then I guess I could reduce the resistance. I did use heat sinks for the Q3 and Q4 - they seemed to fit okay.

I am fairly new to the club so I am not sure about the history on this board. I certainly appreciate what you have uncovered so far. If you don't mind keep me posted.  Hopefully we will get some more responses.

Harold Rothwell

On Wednesday, December 17, 2014 10:41:38 PM UTC-5, Bob Bell wrote:
Sorry if this is already posted , but I searched and could not find anything:

1.  I have started building this board as the basis for a second S-100 system to be used mainly for testing and development work on new S-100 hardware.  This will keep my original S-100 machine up and running for other things.  The question: as I was rounding up the components, I discovered that some don't go the whole way through this overly thick board.  For example, the 8-pin IC socket (a high-quality AMP duel-wipe socket - the only ones I use) and the 2 & 3-pin jumpers.  So if they don't stick out the back side, and its not possible to solder to component side pads, how does one solder them?

2.  Has anyone made a substitution for R6 and R7, listed as 910R.  That's one value I can't find in my junk box, and Jameco won't sell less than 100 of this value at a time.  1000R is within 10% and I hope it would work.  Anyone try?

3.  Why do the fuses only protect the power indicator LEDs?  I checked this carefully to be sure the schematic was not wrong, but tracing the runs on the board shows that the power comes in on P1, +8V on pin 1, and directly to the buss, pin 1.

4.  The values for the current-limiting resistors in series with each of the LEDs seems to me to be very high.  For example, R16 is shown as 20K.  That would result in a LED current of about 0.7mA.  I think I would have to turn the lights off to see the feeble glow from an LED with this current.  I will be using 2K2 for the +16 and -16, and 330R instead of 4K7 on the +8V LED.

5.  I think I found the pushbutton switch at Mouser.  They have a TL1105 series from E-Switch.  Anyone know if that will work?

6.  At one point, I think I saw a photo of this board with heatsinks on Q3 and Q4.  Yet the version 2 board seems to allow little room for heatsinks of any size.  Can anyone tell me if these actually dissipate much heat in normal operation?

Thanks in advance to a great group!

Bob Bell


--
You received this message because you are subscribed to the Google Groups "N8VEM-S100" group.
To unsubscribe from this group and stop receiving emails from it, send an email to n8vem-s100+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.