• Please review our updated Terms and Rules here

Cromemco JS-1 Joystick Replica

I got my Kraft controller today and had an opportunity to take it apart. As you can see, the joystick mechanism is completely different from the original JS-1. It looks similar to the one cjs posted in #58, but not quite.

I took some readings of the pots for comparison and we may have a small issue.

For both Kraft joysticks and in both directions, I got resistance from ~ 0.71K to ~1.4K

For the JS-1 joystick, in one direction I get the exact same reading. However, in the other direction I get readings of 0.84K to 1.62K.

PXL_20240418_191133087.MP.jpg
 
DougM said:
Here's what I usually put together before I head down to the shop. Obviously all of these numbers are made up.

Once I have the thing programmed into the machine I can cut a sample faceplate and you can test fit before I actually try to cut the box. That might avoid $34 mistakes.View attachment 1278102
Click to expand...

Hi,

I like to laser cut cereal boxes or cardboard for test pieces first. That way I can set the piece over an original item, see how it fits before thinking of making the new piece.



.
 
nullvalue said:
Do you think this 2x3" speaker would work? I already ordered them because they were a total bargain, and were the same brand/vintage of the original: https://www.ebay.com/itm/145709281294 - they are 8ohm, I don't know the wattage though.

Also, would a small PCB work better than a tag strip? I wouldn't mind designing one, once you finalize the schematic.
Click to expand...
Yes, those 2x3 inch speakers would work fine.

A pcb would be a little more tidy than the Tag strip they used. But it could be done just by adding another tag strip and leaving it mainly original.
 
nullvalue said:
I got my Kraft controller today and had an opportunity to take it apart. As you can see, the joystick mechanism is completely different from the original JS-1. It looks similar to the one cjs posted in #58, but not quite.

I took some readings of the pots for comparison and we may have a small issue.

For both Kraft joysticks and in both directions, I got resistance from ~ 0.71K to ~1.4K

For the JS-1 joystick, in one direction I get the exact same reading. However, in the other direction I get readings of 0.84K to 1.62K.

View attachment 1278163
Click to expand...
The exact value of the pots should not be a problem, in that they are strung across the zener regulated +10 and -12V power supply rail and their output voltage merely encodes the angle. Their output feeds into a high input impedance system, so it should make little/no difference what very the exact value is. Measure between the pot's two outer terminals and ignore the slider.

Though there is always the possibility that different pots have a different rotational angle that the resistance is distributed across, but most are similar. For an exact test you could apply say 10V across the Pot's outer terminals, and check the voltage on the slider with a full range of movement of the Joystick arm and compare the two. (don't apply the power source to the pot's slider or central pin, only connect the voltmeter there for measurements). I will do this today and get the values for the ones in my Kraft unit.

I would think anything in the range of 2k to 10k pots would work.

Much lower than 2k (1k total, 2k per pot) the current could climb so that the 10 & 12V zeners failed to regulate due to excessive voltage drop across the 220R Ohm resistors. With 1k total, the current would be 22v/1000= 22mA and the voltage drop across the 220R resistors = 4.84v, so the 12V zener would still be regulating as the applied voltage would be 18-4.84v =13.16v.

Much higher than 10k, some current drawn from the Pot's slider could affect the angle to voltage output relationship.


It appears in the Kraft Joystick pots in my unit are 5k, at least that is what their label says, which would appear to be a perfect value, but I don't know for sure if all the Kraft transmitters used 5k pots.
 
Last edited:
DougM said:
You are correct that I can only go as deep into corners as the diameter of my cutter, which will either be 0.125 or maybe 0.200 depending on whether the aluminum loads up. I always cry a little when I snap a 125 cutter.
Click to expand...
The local engraving store told me that the smallest cutter they can use is 1.5mm diameter which is about 0.06". Apparently any smaller than that they have a penchant for breaking.

A while back they did a project for me, so I could make a replica enclosure for another project, there are pictures in here of the panels they cut out with the 1.5mm tool on pages 28 & 29 of the article:

 
DougM said:
Do we have a similar drawing for the joystick?
Click to expand...
No but with using the Kraft enclosure I was able to take some measurements.. let me know if I missed anything else you'd need:

Main hole is ⌀ 1 5/8".

4 mounting holes are each ⌀ 1/8"
centers of opposite mounting holes are spaced 1.850" apart

Distance from center of right mounting to start of rectangular opening is 3/16"

Rectangular opening is 1" x 1/8"

PXL_20240419_001613335.MP~2.jpg
 
Here are the DXF's in case anyone wants to play around with them. I had to zip them because this platform won't accept native DXF


I don't have a laser cutter in house but if @wperko If you want to laser cut them and send to @nullvalue for confirmation that might be good.

Next up, what is the size of the face plate portion of the enclosure?

​

 

Attachments

  • Cromemco DXF's.zip
    8.6 KB · Views: 3
Last edited:
Hugo Holden said:
Apart from the size of the square holes and the Kraft hole pattern, how did we figure out the exact placement of all the holes on the front panel ?
Click to expand...

I don't think we have yet :) I was waiting until I got the enclosure to see if the joystick/gimbal needs to be moved back any due to height restrictions inside this slightly shorter case. Did you get your enclosures yet? Mine haven't even shipped. Was the joystick inside your Kraft controller the same as mine that I posted?

BTW if anyone else is thinking about building these, there are some dual-stick Kraft controller auction listings on eBay currently. Please let me know if you intend to bid on any of these, otherwise I was going to bid on one or two just to have some extras. I just don't want to end up bidding against anyone here.

www.ebay.com

Vintage RC Kraft Transmitter Series Seventy One Radio Controller Dual Stick | eBay

The Radio needs a good cleaning--not working.
www.ebay.com
www.ebay.com

Vintage RC Kraft Transmitter Series Seventy One Radio Controller Dual Stick | eBay

Find many great new & used options and get the best deals for Vintage RC Kraft Transmitter Series Seventy One Radio Controller Dual Stick at the best online prices at eBay! Free shipping for many products!
www.ebay.com
www.ebay.com

Vintage RC Kraft Transmitter Series Seventy Radio Controller Dual Stick | eBay

Find many great new & used options and get the best deals for Vintage RC Kraft Transmitter Series Seventy Radio Controller Dual Stick at the best online prices at eBay! Free shipping for many products!
www.ebay.com
www.ebay.com

Vintage RC Futaba Transmitter Radio | eBay

The radio is a little stained and dusty.
www.ebay.com
www.ebay.com

Vintage RC Futaba Transmitter Radio with matching receiver FPR6F | eBay

The radio is a little stained and dusty.
www.ebay.com
 
Last edited:
Hi,

As another alternative for those of us that look for function over reproduction ... I built two of David Hansel's DAZZLER boards for the Altair-Duino computers. The joystick for that project is designed with thumb joysticks;
https://www.hackster.io/david-hanse...ltair-simulator-3febc6#toc-joystick-support-4

The size of these joysticks works well even with my large hands, and they're cheap to build, although there is no speaker in them, but a speaker box could be made separate from the joysticks on the cable from the Cromemco D + 7A board ...

1713566532833.png


.
 
I have examined the Joysticks in my Kraft unit.

Interesting, they appeared to have moved to a sealed type quality potentiometer (see picture , probably a Bournes part) I would guess that the open frame type with the phenolic wafer became intermittent over time.

In any case I was surprised to find there are two ways in which the potentiometer can be set (before any manipulation of the joystick). One is the entire pot carrier, ,it can be rotated (that is fairly stiff and the pot stays put where it is left). The other is the small thumbwheel knob.

I applied 10 V to it for a test, surprisingly the output voltage was 6.22v. But Then I discovered the ability to rotate the whole pot to a new position. So with the thumbwheel in the center & joystick centralized I rotated the pot to give 5V output.

Then, the thumbwheel full defection each way alters to voltage to 5.2 or 4.8V so that is a +/- 0.2 V deflection.

The joystick itself produced +6.15 V in one direction and +3.86 in the other direction. So about +/- 1.14V deflection.

So if it was running off 22V as in the Dazzler control box, it will give pretty well exactly bang on +/- 2.5 volts. And the Thumbwheel would give +/- 0.4v.

It would be interesting now to measure the original Joystick.

To double check this I applied 22V , with the Joystick centered, the voltage was 11V, going to +13.5V in one direction and 8.5V in the other , so +/-2.5V
 

Attachments

  • jstick.jpg
    jstick.jpg
    163.3 KB · Views: 5
  • Like
Reactions: cjs
Hugo Holden said:
It would be interesting now to measure the original Joystick.
Click to expand...

I'd be happy to give this a test with the original, but can you clarify which leads of the pot you applied the voltage to, and where you took the readings from?

Also what thumbnail knob are you referring to? Is this the trimmer adjustment (the 2 rectangular openings on the Kraft controller), or something else on the joystick?
 
Hugo Holden said:
I have examined the Joysticks in my Kraft unit.

Interesting, they appeared to have moved to a sealed type quality potentiometer (see picture , probably a Bournes part) I would guess that the open frame type with the phenolic wafer became intermittent over time.

In any case I was surprised to find there are two ways in which the potentiometer can be set (before any manipulation of the joystick). One is the entire pot carrier, ,it can be rotated (that is fairly stiff and the pot stays put where it is left). The other is the small thumbwheel knob.

I applied 10 V to it for a test, surprisingly the output voltage was 6.22v. But Then I discovered the ability to rotate the whole pot to a new position. So with the thumbwheel in the center & joystick centralized I rotated the pot to give 5V output.

Then, the thumbwheel full defection each way alters to voltage to 5.2 or 4.8V so that is a +/- 0.2 V deflection.

The joystick itself produced +6.15 V in one direction and +3.86 in the other direction. So about +/- 1.14V deflection.

So if it was running off 22V as in the Dazzler control box, it will give pretty well exactly bang on +/- 2.5 volts. And the Thumbwheel would give +/- 0.4v.

It would be interesting now to measure the original Joystick.

To double check this I applied 22V , with the Joystick centered, the voltage was 11V, going to +13.5V in one direction and 8.5V in the other , so +/-2.5V
Click to expand...
An analysis of the voltage swing on the original Cromemco JS-1 is detailed here:

S100 Computers - Dazzler Joystick Board

S100 Computers Cirrus GD5420 VGA board
www.s100computers.com
The JS-1 wiper voltage swing is 4.5V.

Your measurements from the joysticks removed from the Kraft RC unit match pretty closely to that of the original, and are a good match for the +/- 2.55V input voltage range of the D+7A board.
 
Last edited:
nullvalue said:
I'd be happy to give this a test with the original, but can you clarify which leads of the pot you applied the voltage to, and where you took the readings from?

Also what thumbnail knob are you referring to? Is this the trimmer adjustment (the 2 rectangular openings on the Kraft controller), or something else on the joystick?
Click to expand...
Generally a potentiometer has two outer connections on the far end of the track. The central connection is placed symmetrically between those two, called the "slider" or the "wiper" because it is connected to the the moving (rotating) arm that slides over the resistive track.

The test simply involves applying say 10V DC (or 22V in this case) to the two outer connections. This applies the voltage field across the pot's track. If it is a linear potentiometer, then the electrical potential is distributed evenly across the track. Say if 10V was applied, the physical center of the track, midway between the two ends would, on the meter, measure 5V to each end.

Connecting the volt meter's negative lead to one end (usually the negative connection of the supply is customary) and connecting the positive lead of the voltmeter to the central slider connection, to measure the voltage.

It looks to me as though @hmb has already worked it out and the Kraft Joysticks are perfectly compatible with the ones Cromemco used.
 
On the speakers I recommended for the replica, I just discovered they are also stocked at Mouser.

PS: still waiting on my enclosures too. The watched pot never boils.
 

Attachments

  • SPKR.jpg
    SPKR.jpg
    159.5 KB · Views: 1
Last edited:
Hi,

I'm thinking perhaps a good starting point is Gerber files for the Joystick circuit board ... things like the switches and stick connections should probably be female Dupont headers to allow people to build any type of joystick they want with the components being from original part numbers to almost anything else that meets the requirements of the joystick.

Some people want look-alike, while others just want function. With the Gerber files, we can build up a joystick board and add our own choice of buttons/switches, joystick and even a COTS box or 3D printed or CNC cut box.


.
 
wperko said:
Hi,

I'm thinking perhaps a good starting point is Gerber files for the Joystick circuit board ... things like the switches and stick connections should probably be female Dupont headers to allow people to build any type of joystick they want with the components being from original part numbers to almost anything else that meets the requirements of the joystick.

Some people want look-alike, while others just want function. With the Gerber files, we can build up a joystick board and add our own choice of buttons/switches, joystick and even a COTS box or 3D printed or CNC cut box.
Click to expand...

Walt, the goal of this build was to stay as authentic as possible. For the buttons, we're going with the Cherry switches which are panel mount, therefore they won't need to be on a PCB. We may build out a PCB for the other components, not sure yet.
 
Hugo Holden said:
Connecting the volt meter's negative lead to one end (usually the negative connection of the supply is customary) and connecting the positive lead of the voltmeter to the central slider connection, to measure the voltage.
Click to expand...

Here are the readings taken from the original JS-1 @ 22V
Code: 
AXIS            -           CENTER          +
Left/Right    14.51V        16.47V        18.50V
Up/Down       14.28V        16.07V        18.19V
 
hmb said:
An analysis of the voltage swing on the original Cromemco JS-1 is detailed here:

S100 Computers - Dazzler Joystick Board

S100 Computers Cirrus GD5420 VGA board
www.s100computers.com
The JS-1 wiper voltage swing is 4.5V.

Your measurements from the joysticks removed from the Kraft RC unit match pretty closely to that of the original, and are a good match for the +/- 2.55V input voltage range of the D+7A board.
Click to expand...
I'm doing more work on this. According to this document the output voltage from the Joystick (in the original Cromemco unit) should sit on zero volts and defect each way 2V. Or a 4V swing. Not 4.5v.... see Page 5 under Joystick operation:


(Probably, this Cromemco document is unlikely to be incorrect, but stranger things have happened, and the D + 7A board is designed to accept +/- 2.5V so it would seem odd that the full dynamic range wasn't utilized).

I'm not sure why that disagrees with those remarks or analysis on S-100 computers. That assessment implies +/- 2.25V.

So which is it +/- 2.5V, +/- 2.25V or +/- 2V ? Perhaps all close enough to work.

I had also wondered why they used a 12V and 10V zener in the original unit and not simply powered the Joystick pots by +/-10v. I think they needed 22V total because the original Joystick didn't quite get to -/+2 "point something Volts" output just run on 20V, or they would have simply used two 10V zener regulators.

By doing that though with the 10 & 12V supplies, there would be a 1V standing offset on the original unit, which would need to mechanically nulled out by the rotational position of the pot on the joystick body. The thumbwheel on its own would not have enough range.

I think in our replica we should power the the joysticks from +/- 10v, which will give +/- 2.28v from these Kraft Joysticks.
 
You must log in or register to reply here.
Back
Top