• Please review our updated Terms and Rules here

Interested in a home brew Z80 computer project?

I may be interested in one or a couple (2-5) boards. To summarize: are there a lot of components one needs to gather yourself, or do you offer kits of the required parts as well as the boards themselves? I suppose the most basic resistors, diodes, capacitors and so on are easy to find locally, but in case there some more specific items it is good to recapitulate this.

Hi Anders,
All I sell is the PCB. I'll sell the pre-programmed EPROM only if needed. The rest of the parts are up to the builder to find and source. There are only commonly available parts in the design although that may vary depending on where you are. I do not sell kits or other parts to keep costs low and encourage salvaging or other innovative methods of gathering parts to keep costs low.

Thanks for asking! Have a nice day!

Andrew Lynch
 
I may be interested in one or a couple (2-5) boards. To summarize: are there a lot of components one needs to gather yourself, or do you offer kits of the required parts as well as the boards themselves? I suppose the most basic resistors, diodes, capacitors and so on are easy to find locally, but in case there some more specific items it is good to recapitulate this.

Hi Anders,
I thought about your question a bit more and would like to expand a bit on my answer from yesterday.

My goal for the N8VEM SBC project is to encourage and help people start home brew computing. Making an SBC kit actually detracts from that goal in two important ways:

1. part selection and acquisition are key skills necessary for the home brew hobbyist for building your own computers or any other electronics project. Doing so builds necessary skills and gives insight into the design building a kit never will. Also few things teach better than making some mistakes -- especially when desoldering is involved. This is probably the most important reason against making a full blown SBC kit. It basically robs the hobbyist of much of the fun and benefit of doing it in the first place.

2. investment and risk to make full blown kits is MUCH higher than just a lot of manufactured PCBs. The manufacturing of PCBs alone cost me hundreds of dollars which is a significant investment for a regular working person. Adding to that lot buys of components would increase my investment and risk several fold. If those kits don't sell, I am out some serious funds and will have to explain that to the family. This practical limitation is probably this is why so few hobbyist kit projects exist. Those which do exist tend to be small and inexpensive. Very few people have thousands of dollars available to risk on a non-profit venture with an unknown customer base and unknown demand. Worse, the risk of being straddled with unsold kits increases with the cost of the kit -- the more expensive it is the less likely it is to sell. Frankly, I just can't stand the pain of losing that sort of family money and I doubt many others could either.

Hopefully this gives some insight into my reasoning on the home brew vs. kit decision.

Thanks and have a nice day!

Andrew Lynch

PS, if anyone is interested in buying a PCB, please contact me so you can get on the waiting list for the next batch.
 
Thank you. I will look up later exactly what the parts list looks like and determine how many components I will be able to find without spending a small fortune. As for EPROMs, both me and my friends (for whom the 2-5 boards would go to) have our own programmers, only the chips themselves would be required to obtain. I saw from previous posts in this thread places to order components from, but those places usually only make sense to people in the North America, even though the USD still is quite low in value.
 
Hi Anders,

On the N8VEM Google Group there are a couple of "parts list" documents. I wrote one using Jameco and Howard wrote one for DigiKey. Those are both usable but reflect different perspectives and approaches and will result in very different final total parts costs.

If you'd like they are available to help review and guide personal selections. Of course, discussion on the costs and benefits before soldering starts or even ordering parts is good preparation. You should be able to order parts locally and/or reuse parts available from salvage, etc. Certainly there are quite a few European parts suppliers with more than adequate selection.

The whole parts selection process has lead to some very excellent discussion threads as builders dig into what is required and their own design decisions. Overall this is exactly the purpose of leaving the parts list undefined is so the builders go through the process of evaluating parts and performance vs. cost trade offs.

It's my guess that every SBC which gets built will be unique and reflect the personal choices of the builder. Exactly as it should be for home brew computers -- they are all unique.

Thanks and have a nice day!

Andrew Lynch

PS, having at least one EPROM programmer among you and your group of friends should be sufficient for updating the contents of the ROM drive and/or system programming. Usually burning your own EPROM is not something you do everyday but when you need it, you'll really enjoy having one available. There is already a builder who has started rewriting the CBIOS which I very much appreciate. No doubt, the current CBIOS could use some review and new ideas!
 
Hm, ok. So the values of many common components are defined within some ranges, i.e. 300-700 Ohm, 100 - 1000 nF, any diode that will prevent back current at 1V and so on? That's cool, and I understand it will lead to some discussion which parts are better suited than others if you dig deep into the specs.

As I wrote, I will have a look and ask around if we're interested to be put on the waiting list. There is a group of us with highly varying soldering and building skills who have got together for a little over a year, sitting down to putting together a common project. Last year we built the ZX Spectrum divIDE interface from a kit: four out of seven finished soldering, two out of four got it working and one out of two managed to avoid getting the interface die without reason a few months later. This spring we built a Velleman Pong kit which was much simpler than the divIDE interface, and I think everyone succeeded although it was not nearly as fun - some people had ideas about further modifications but I don't think anyone has done anything to it.

A SBC Z80 computer would be an interesting, challenging project in line with my requests that we find something that is not about assembling straight out of a box, but my friends may disagree, in particular if the costs get too high.

There are well-stocked electronics shops here, like ELFA, but they tend to be a lot more expensive than your Digikey or Jameco, even if international shipping was considered. Some other stores sell components too, either as individual items or in large bags of assorted surplus. The latter is really affordable, but you never know what's inside the bag until you get home. For those who already have a big box of most typical components it is not a problem to get a few missing pieces, but if you virtually has nothing to start with you got to decide more carefully what you buy. For example, I'm fond of wiring resistors in series to reach a higher resistance, something I might do on a N8VEM board too if I didn't have resistors of suitable values.
 
Hi Anders,

Your hobby electronics group certainly sounds like the N8VEM SBC project would be right for them especially if this is a group build project. The members on the N8VEM website are of varying experience levels but that doesn't seem to be causing any trouble. My target audience for the project is electronics and computer hobbyists with education and entertainment in mind. I'd like the project to be challenging and educational but not so hard that new builders are discouraged.

Actually ordering parts for a group may be easier and you'll get a better value if you can combine the orders. I know Jameco and others give discounts for parts based on quantity. Also, some parts are sold in minimum quantity like resistors so it would actually be beneficial to order a 100 resistors and split the order among the group to share costs.

All of the parts are very common items so the price should not be too bad but if you are considering a large group order of parts, I strongly recommend a thorough review of the parts selection and even asking for an outside review. Having to do follow on buys for limited parts will wreck any economy of scale.

However, I don't know your group so you'll have to be the judge as to whether this is appropriate for them or not. During the design phase, I really emphasized to make the PCB easy to assemble by maximizing the clearances to reduce bridging. Adding the solder mask really seemed to help too. I haven't heard of a single mention of solder bridging since I built the initial PCB prototypes which had no solder mask and only minimal clearances.

That being said though, I don't have an estimated date for a second batch order of PCBs. It will depend on demand as to whether and when that will happen. Probably I will let things settle for a while and help the builders complete some projects. It seems that word and pictures of successful builds is the best sales pitch rather than me trying to explain the project.

I hope this helps. Please let me know what I can do. Thanks and have a nice day!

Andrew Lynch
 
Yep, I just posted to the major Swedish vintage gaming/computing site. Let's see if I spun some interest. In the best case, I may end up with putting an order of 10+ boards on the waiting list?
 
Yep, I just posted to the major Swedish vintage gaming/computing site. Let's see if I spun some interest. In the best case, I may end up with putting an order of 10+ boards on the waiting list?

Hi Anders,
Wow! If you can get a reasonably firm commitment to buy 10 PCBs we don't need a waiting list. I will just order a new batch straight away. All I am looking to do is have a semi-reasonable chance of covering my expenses.

Please let me know what comes of it. Thanks! I do appreciate your initiative!

Andrew Lynch
 
Hi All,
It looks like another N8VEM SBC has sprung to life with photos to document the joyous occasion! Congratulations Juha! Excellent work!

http://n8vem.googlegroups.com/web/a...le2G1qiJ7UbTIup-M2XPURDS0mIZv3KyzgLe-Efpubu6n

Check the N8VEM Google Group site for more details.

This would make it three SBCs in a row working for hobbyists other than me. Definitely a good sign! Each one is unique and highly personalized.

Things are looking up!

Thanks and have a nice day!

Andrew Lynch
 
Yep, I just posted to the major Swedish vintage gaming/computing site. Let's see if I spun some interest. In the best case, I may end up with putting an order of 10+ boards on the waiting list?

Hi Anders,

Has there been any interest in your post? There have been a few people joining the waiting list. I think I am up to about 7 people now so the I am getting close to ordering the second batch of PCBs.

Probably only about 3-5 more would be enough. 10 is the working minimum but it'd be nice to have a couple of backups in case people can't due to circumstances.

Please let me know how things are going. Thanks and have a nice day!

Andrew Lynch
 
Hi!
Is anyone still interested in this subject who is not already involved in some way?

Thanks!

Andrew Lynch
 
Hi! If anyone would like a N8VEM Z80 SBC PCB, the second batch of PCBs has arrived and I am ready to process any orders. The first batch went fairly quickly so I bought more.

All the orders on the waiting list have been sent already and there are plenty PCBs left for anyone who would like to build their own home brew Z80 CP/M SBC from scratch.

The good news for those who waited for the second batch is that there are several successful builders who have published photos of their working SBCs. I think it is safe to say now you don't have to worry this SBC design may have contained some fatal design flaw.

You'll benefit from the first group which went before and has posted lots of information on how to make your project a success. The SBC has been demonstrated working numerous times by many kinds of builders of varying degrees of experience.

Please send me an email if you are interested! Thanks and have a nice day!

Andrew Lynch

http://groups.google.com/group/n8vem
 
N8VEM SBC finally gets its BlinkenLights!

N8VEM SBC finally gets its BlinkenLights!

Hi,

I recall someone in this thread a while back asking about the N8VEM SBC and saying they were looking forward to seeing its "BlinkenLights". Well, that day has finally come!

http://n8vem.googlegroups.com/web/IMG_6726.JPG

This afternoon I built up an ECB backplane and got the N8VEM SBC to work with it. The very first N8VEM ECB peripheral is my old TestPrototype Bus Interface Test Tool sometimes referred to as the Bus Interface Test Checker (the BITC) ;-)

The good news is the SBC and ECB backplane have all checked out and the BITC was helpful in getting it all to work together. Although it was cranky about it as usual and not terribly reliable, the BITC is one of those tools I used a lot when doing the same types of testing on the TestPrototype system.

It is a tremendously simple peripheral and practically useless but it demonstrates the general principles behind making a simple ECB peripheral. All it has is some simple IO decoding logic, an intel 8255, and a series of LEDs to indicate state of one of the parallel outputs. I also added LED status indicators for the various bus pins like Address, Data, and some Control signals.

You can watch the N8VEM SBC go through its boot sequence and you can generally tell what it is doing based on the LED patterns. All the i8255 does is demonstrate the IO port logic is working. Once that is in place any Z80 peripheral should work *in theory*.

Thanks and have a nice day!

Andrew Lynch
 
I have a multiuser(2) z80 cp/m based single board system here that I built decades ago for a video industry design.
As it was a "manufactured device" it has artworks for the pcb and I still have all the details of the associated bits and pieces (epld programs, monitor rom, cp/m source etc).
It now uses an IDE hard disk because the original MFM drive died and it is not easy to get a replacement and there are trillions of old ide drives around (well maybe I exaggerate here but ide is very easy to implement and there are probably billions of them floating around).
If you think any of this may be useful to a/the project concept then let me know and I can post photos on my web site.
 
Hi!

Yes, please do post you photos & documentation. I am curious as to what your SBC design is.

You say you designed your own Z80 SBC? Are you interested in a little more Z80 SBC design? How are you with peripheral development? The N8VEM project is getting ready to start the peripheral development phase.

This weekend I built and tested the ECB backplane with the SBC and it is working. My TestPrototype BITC and Disk IO boards are working so I have primitive bus monitoring and the IDE disk interface working. The FDC section of the Disk IO board still needs to be tested, etc but signs so far are hopeful. The new PCBs seem to be much more reliable and the theory about improved grounding seems to be holding up.

I would certainly appreciate a hand in the design area. If nothing else, you could do some independent review of my upcoming designs to make sure they are as error free as possible. Specifically the ECB bus monitor (BITC replacement) is in the final phase and needs some fresh review before I send it off for manufactured PCBs.

Do you do any prototype development? (wire-wrap or point to point soldering on prototype boards?)

Thanks and have a nice day! Please consider joining us on the N8VEM Google Group.

Andrew Lynch
 
z80 sbc

z80 sbc

Hi again,

I have posted some photos on http://www.candl.com.au/cpm today.
You can work out the architecture from the chip numbers but basically it is a 10mhz z80 with 128kb nvram (64k for each user) and 4k nvrom for the monitor.
There is a mfm disk controller, now not used, bottom left, which if not installed makes the board 1/2 the size.
The ide add-on is top left and is the best bit about the system as it make it worth keeping. The floppy is a 1.44MB pc floppy formatted for 720k to be compatible with my original prototype. That system was wire wrap but is long gone I now breadboard stuff and then solder a final version (see ide add-on) when finalized.
The pcb in front is the hsc inc 68k co-processor which is altogether another story.
I will post all the electronic bits I can find over the weekend as it will take some time to find it all.
You are welcome to use any/all/none of what you find, however, if you find it useful please add a line in any derived stuff stating where it came from (so I can apologize for any inconvenience caused)
This board was designed for Glaxo Australia to implement a voice alarm system back in the early 90's and was used there of well over a decade, only retired when the mfm drives failed (hence the ide add-on (which never went into production (sadly)))
 
Hi!

WOW! JUST WOW!

That is one cool SBC! You made an ST-506/ST-412 MFM interface? I wouldn't even consider it as it is WAY over my head. I am impressed!

Did the system have on board video? I see it uses the WD FDC2793. Sweet FDC but it sure is a shame practically every recent PC uses the not so nice NEC765 and derivatives.

I'd like to ask about the 68K but I can see that is probably quite a story. Are you familiar with the ECB bus? There was some interest in making 68K home brew systems for ECB. I have seen bits and pieces around but they are pretty rare these days. I am so swamped right now I cannot even consider the possibilities until I get some of these projects finished.

There are two N8VEM ECB peripherals in work right now that could definitely use your help. The N8VEM SBC and ECB backplanes are both done and working so it is too late for those. :) The PCBs are available for home brew hobbyists but I would like to transition the project into the peripheral development phase. It's easier said than done though. I am trying to recruit some talented individuals and you just joined the short list.

I am making an ECB bus monitor and am almost done. The schematic and PCB layout is in final review now but I could use some rigorous peer review as to whether this thing will actually work and if there are any improvements I could roll in -- preferrably while preserving the original intent of the design. The original design is derived from a German ECB project in C'T Projekt magazine and follows it fairly well but I have made some extensions and modernized it. The peripheral operates somewhat like a Jade Bus Probe but for ECB bus rather than S-100. It's sole purpose is for hardware/software development but could be used for other Z80 type projects since ECB is so similar to the Z80 bus structure. Here is a link -- caution 29MB download!

http://www.hd64180-ecb.de/assets/download2008/ct_bus_monitor.pdf

The other peripheral is the "crown jewel" of the project at least in my mind. It is the Disk IO board which contains an IDE and an FDC interface. The IDE interface is known working and I have a CP/M CBIOS which supports it. I have my prototype version running in my basement workshop and I am fairly confident it is OK. Here is a link from the N8VEM website. You can barely see the circuit due to all the wiring and it is a relatively low complexity board.

http://n8vem.googlegroups.com/web/IMG_6757.JPG

The FDC section is another story though; my original prototype started getting flaky last year due to poor grounding, I think. I switched the project to all manufactured PCBs to address the poor grounding of the prototype boards and improve reliability. So far, that has paid off nicely and the SBC and ECB backplanes do not exhibit the constant flakiness the original prototypes did. Even my prototype peripherals just work better. However, I halted the project after building the FDC but never tested it. I have some simple FDC software written but again never even tried it out. :-(

Now I am too busy scrambling to get this ECB bus monitor finished and managing the N8VEM project to work on the FDC. My plan is to finish the ECB bus monitor and then start on the testing of the FDC portion of the Disk IO board. Once I know it works and/or rolled in the fixes then I'll convert it to manufactured PCBs like the rest of the project.

So, are you interested in reviewing a couple of designs? The ECB bus monitor is straight TTL chip logic and the pieces are not all that complex. Still the overall function is rather complex due to its use of one-shots, etc. The FDC section on the Disk IO board could use some TLC as well. How are you at building prototypes? :)

Please consider joining us on the N8VEM project. Seriously, I could use your help. Thanks!

Have a nice day!

Andrew Lynch

http://groups.google.com/group/n8vem?hl=en
 
Hi Andrew
I will look at the links you have posted although I will be going from scratch as I have only come across this forum this week.
I like things to be "the smaller the better" hence the system I have here. I probably wouldn't build a bus oriented system for myself as integration makes it nicer to hide hardware. (I am thinking I will be putting the board I have here in my PC case and getting rid of the horrid box on my desk. I would like to try sharing the ide drive on the PC with the CP/M one just to see what I get) This may mean I can have the z80 system AND the PC accessing one disk although I suspect not.
This system is serial console only as a typical cp/m system originally was. Also being 2 user serial is lots easier to implement. I would like to redo the board with the mfm stuff replaced by the ide circuit and shrunk to maybe half its size.
The 68K board is not really required now the ide is running so I might remove it.
I have posted the pcb layouts and schematics today and will find the monitor source/binaries and a floppy disk image for the cp/m development system this weekend. This makes it all usable rather easily.
It might be worthwhile making one of these up for yourself to augment what you are doing as I can give you a complete floppy or ide based image from what I have here although you would only have access to the duarts I use peripheral wise. Let me know if you or anyone else decides to do that as the ide circuit isn't drawn up except as a sketch.
I guess you know the software apps I have posted (autotrax and schedit) as I understand they are common public domain systems from last century. The aplus stuff I am not so sure about but as the jedec files are there you dont really need to even run it. You can do whatever you like with these pcb's and schematics if you can do things cheaply there overseas. I have no idea how you make boards so cheaply there in the US as I paid hundreds for the 4 I originally produced here in Australia.
Without the mfm controller you can power this thing from +5v only.
Note it was nearly 20 years ago I did this z80 design and then went down scale, processor wise, from there (motorola embedded devices hc5's and dallas ds5000's) so I have not though about this stuff for a long time, but it might be good to exercise my mind again.
Thinking as I type I reckon putting this system on a PCI card to derive power and plugging it into a pc would be a neat way of hiding it whilst keeping actual z80 systems alive!
Anyway I will visit the project over the weekend and have a look.
 
re N8VEM

re N8VEM

I had a browse of what you are doing there and it is quite impressive. You look like you have come a long way.
For a bus monitor tho' would'nt it be better to use a PIC based controller to provide an lcd display and single-step/trace facility using busrq/halt rather than a discrete system.
Although, what I do here and used way back when, is an Arium ML4100 with a z80 pod and zsid for debugging although old hardware is getting more scarce as it gets junked (ebay is still worth a look regularly if you live in the States).
Unfortunately I am not in a position to built a whole new bus oriented system as I have no bench space in my little room here, so will keep and maybe one day shrink my existing system, so I dont think I can be of much help.
I am slowly putting up all the info on my system on the web site http://candl.com.au/cpm/ for easy access so if any of that info is helpful to you or anyone else you are welcome to use any of it.
Also after playing with the z80 system the last few days I decided to separate the hsc 68k board and re-engineer it to use some old 4mb 30p simms I have laying around to give it more memory (it only has 2m) and put a duart on it so I can run it standalone.
I will then junk my old rack and mount the z80 (and hopefully the 68k as well) in my existing PC case to hide the electronics and just keep it all for posterity.
If you have any questions or would like me to look at something you are doing from a "theoretical" perspective you are welcome to pm me.
 
Back
Top