how to lighten BS springs?

[wellington1869]

Ok, here's a question.

Seeing the switch poll in the other thread, I was suprised that blue cherries are so far ahead. Seems to me that with each passing year the old IBM BS system will have fewer and fewer fans. Since a lot of the fandom there is driven by nostalgia, I have to admit.  For modern users, probably BS are really too stiff (even tho they're not stiff in comparison to what old folks are used to, typewriters and older keyboards, where something around 60g of resistance seems to be the norm). These days if you're over 45g of resistance people describe it as stiff.

So, question:

Is there a way to lighten the BS spring?
Possible solutions:
a) find a replacement spring with less tension? This may be difficult unless custom built (and so prolly expensive).

b) Would the springs lighten if, say, stick the keyb under a coffee table for two weeks with all the keys in a depressed state?

c) Here's a radical idea: get some relatively mild metal-eating acid. soak the BS springs in it for short periods of time. Would this "thin out" the springs just enough to reduce resistance by, say, 10g? (around 50g resistiance might be ideal for the "BS-light").

Any other ideas or comments on the above? How to get from point A to point B in this case?

[itlnstln]

If there were a lighter BS, I would be all over it (that is, if it were around <= 50g). I really like buckling springs, but as Wellington said, some find them too stiff, and I do.
 
I wonder if Unicomp could help...

[wellington1869]

I wonder how much it would cost to approach a spring-maker and give them a bs sample and say "reproduce this but lighten this to 50g and we'll take 1000 springs".

[wellington1869]

I wonder if (b) would work. Anyone want to try it? Use rip-O-meter before and after the coffee table.

[talis]

Changing the spring may ultimately result in not enough pressure being put on the membrane by the foot when the spring does buckle, and may result in missed keystrokes.  At least on a model M, a model F is more likely to keep working properly even with a spring change.

You'd also likely loose some of the sound, and the tactile feel with a softer spring.

Its possible that it will work, but I suspect that IBM designed the keys with that weight for a reason (since it seems its heavier then some of their earlier typewriter designs).

[ch_123]

I must get my hands on this dry teflon stuff and try using it on my Model F. Even though it's lighter than a regular BS board, I'd like it to be even lighter still.

[wellington1869]

can someone do rip-O-meter on the F? Is it really lighter than M?  By at least 10g?

And if its lighter by 10g or so, has anyone replaced M springs with F springs?

Edit:  Caught up with the converstaion on the other thread. Apparently you'd need a longer barrel on the M to just swap springs.

But -- since its the longer spring of the F that is creating a softer feel -- what if:

d) You cut the M spring by a quarter inch, then stretch it to make it longer.  Would that equal a softer M?

[wellington1869]

Here's a list of spring manufactures sorted by County and State.

The specs for the IBM Model M springs are kinda here but don't quote me on the wire diameter part.

its funny, i used to live right next door to a small factory that had a sign that said something like "manufacturer of springs and coils" and I always thought it was funny super-niche market. [after all, someone is making those ball point pen springs!]. Of course if I were still living there today I'd just run over with Bs spring in hand

[wellington1869]

You'd also likely loose some of the sound, .

might not be a bad thing for modern users. Though i wouldnt want the bs to become totally blue-cherry like.

[wellington1869]

Your mind follows strange paths.

I still like my metal-eating-acid-bath idea too

[itlnstln]

I'll try that sometime. While doing the rivet mod I managed to bend a spring over the barrel. It was so slight I didn't think it would matter but a slight bend basically rendered the spring useless - it wouldn't buckle! I ended up swapping springs using Sandy's method that you posted - I took one from my son't Unicomp ScrollLock. Those suckers are easy to put in but hard to pull out without bending.

Another abuse for the Boscom.

[wellington1869]

Also,
e) out of curiosity, how about putting in springs from other keyboards (like a black alps board) by stretching those springs? How would it feel? (similarly, surely there is a cheap ballpoint pen out there that has thin diameter springs that we could stretch out?). Or similarly, surely a manufacturer of springs and coils would have a "catalog" of standard springs from which we could choose one that comes close to what we're looking for? (Stretching it a bit if necessary).

[timw4mail]

I never thought about how complex the mechanism really is...

[ch_123]

can someone do rip-O-meter on the F? Is it really lighter than M?  By at least 10g?

I did it, it's in the 60-65g region. It's hard to get an accurate reading on BS keys, also, spring wear could lower the resistance. Hard to tell which figure is more accurate.

Edit:  Caught up with the converstaion on the other thread. Apparently you'd need a longer barrel on the M to just swap springs.

That's entirely an estimation on my behalf, I figure that the barrel is sized so that there's a snug fit between keycap and spring. If anyone wants to try, they could probably get their hands on a cheap XT Model F, and try swapping with a Model M.

But -- since its the longer spring of the F that is creating a softer feel -- what if:

d) You cut the M spring by a quarter inch, then stretch it to make it longer.  Would that equal a softer M?

The thing about the F is that it's not just a soft Model M, the tactility and bounce-back of the keyswitch is far more pronounced. I think if you did that that you'd just end with a mushy feeling Model M. They offer a completely different typing experience, as Webwit once said - they're like Blue Cherrys done right.

Also,
e) out of curiosity, how about putting in springs from other keyboards (like a black alps board) by stretching those springs? How would it feel?

Once upon a time, before I abandoned the UK ISO layout, I had a worn out Dell AT102W. I got one of those infamous $5 NIB AT101Ws and transferred the switches from one keyboard to the other. Once I had finished, I was playing around with the old worn out ones, I accidentally bent one spring. I tried to fix it by streching it out, but it wouldnt go back to it's old shape when I pushed it back in, instead, I ended up with something that looked like two Topre springs stuck together at the wide end. I put this into a switch and assembled it into the spare PCB. It had an interesting feel, it was stiff and linear. I tried to recreate it, but the second one didnt turn out right, probably one of those fluke chances.

[wellington1869]

thanks for the info CH. IIRC, the M's are about 65-70g, so I guess the F's are about 5g or so lighter then. I'd like to bring them down to 50g

[timw4mail]

thanks for the info CH. IIRC, the M's are about 65-70g, so I guess the F's are about 5g or so lighter then. I'd like to bring them down to 50g

My only real issue with the spring strength is the tendency to make one's fingers bounce too much in normal typing. What are the linear Cherry's force rating? That seems to be a pretty good strength.

(Of course that might be 50g, I'm not sure.)

[ch_123]

My only real issue with the spring strength is the tendency to make one's fingers bounce too much in normal typing. What are the linear Cherry's force rating? That seems to be a pretty good strength.

(Of course that might be 50g, I'm not sure.)

Linear Cherrys are about 60g if memory serves me correct. From my experience with old 1970s and early 1980s linears, the lighter the better. They're rather strange to type on, you'd get used to it after a while, but I'd say you'd miss the tactility.

[wellington1869]

Blue Cherries are 50g.  Not too stiff, not too light.  Might have something to do with it leading in the poll.

are you sure about that? Crap, maybe i need to try blue cherries again, lol.

[Rajagra]

Is there a way to lighten the BS spring?

How about adding weights into the keycaps? Should give an interesting feel!

[wellington1869]

How about adding weights into the keycaps? Should give an interesting feel!

thats an interesting idea. tho I think while downstroke may get easier upward bounce would suffer too. Would be interesting to try it though.

I sort of had this idea when I put what I think are the heavier M keys and stems onto my endurapro, tho the weight difference couldnt have been that much really. It did affect tonal sound though (the thicker old plastics have a lower tone, which I like).

They do sell this epoxy-like weighting glue (I think they use them on model airplanes). Would be trivial to pour some of that into the underside of the keycap, if someone wanted to try it with an old board.

[talis]

Thanks to this thread (and an upcoming mod) I have a bucket of springs and keys to play around with.  I'll see what I can come up with to test some of these ideas.

[wellington1869]

Thanks to this thread (and an upcoming mod) I have a bucket of springs and keys to play around with.  I'll see what I can come up with to test some of these ideas.

that would be awesome talis

[itlnstln]

Cherry blacks start at 40g and finish at 80g; they are a pretty heavy switch. I think spring bounce doesn't have as much to do with key spring/resistance as one might think since it seems rather logical. All the Alps keyboards I have/had do not spring back after the keystroke like all of my Cherry brown keyboards do, and the Cherry brown is a significantly lighter switch. This was one of the reasons why Chloe was such a Cherry fan. When you start to learn how to type light enough, the Cherrys help you out by pushing your finger back up off the key. The Alps only seemed to have enough force to push the key back up and that was about it. There has to be something with the materials used in the spring, the way the spring is coiled, and maybe the weight of the stem of the switch that also determines how bouncy a switch feels.

[ch_123]

When you start to learn how to type light enough, the Cherrys help you out by pushing your finger back up off the key. The Alps only seemed to have enough force to push the key back up and that was about it.

QFT. I can type fastest on my blue Cherry board.

For any Topre users out there - how much bounce-back is there in those switches?

[itlnstln]

Yep, that's what I'm feeling on this AT101W - it makes it feel at least as much work to type as a Buckling Spring if not more.

Even though the Alps switches are lighter than BS, this phenomenon give them the perception (and, perhaps, the reality) that they require more work than BS.

[wellington1869]

Cherry blacks start at 40g and finish at 80g

"like a hard bit of jello" -- chloe

[itlnstln]

I miss Chloe and Sandy.

[ch_123]

According to Sandy's home page, the 5576-A01, which is generally considered to be stiffer than the Model M, has shorter springs. Seems to back up this theory further.

Ripster - how much do those coins weigh again?

[itlnstln]

Ripster - how much do those coins weigh again?

5g each.

[ch_123]

I count 16... So you managed to raise the resistance to 80g?

[timw4mail]

I count 16... So you managed to raise the resistance to 80g?

That's what it looks like. Apparently, you need springs of the same diameter and length with more coils for a weaker spring...weird.

[ch_123]

Yeah, the M and F springs are definitely the same diameter - there's just more 'spring' on the F.

[wellington1869]

lol, thanks for the experiment, rippy!  80g, eh? hmmmm.
I'd stick mine under the coffee table but I only have one working M and I dont want to ruin it!  Someone who has some spare boards will have to try....

Just as soon as you think you understand the Buckling Spring the thing gets all random on you

god thats the truth (or: GTTT).

[wellington1869]

its a real conundrum.  I still think we might be able to find a replacement lighter spring from a standard spring-parts catalog from a manufacturer. maybe a google search is in order...

[timw4mail]

What do I do to make it softer?? No acid spring bath for me nosiree..... End up looking like Phantom of the Opera.

I think you have to get "denser" springs

[wellington1869]

anyone with a spring in hand want to post diameter and length and a closeup pic of the ends?  I can search google once I have the info... (of course i could pull one out of my M but as I say its working fine and I'd rather not mess with a single working board...). If anyone has any broken or spare boards on the other hand maybe you wouldnt mind yanking a spring out and letting us know what to look for...

[JBert]

Your mind follows strange paths.  I was thinking of a blowtorch on the springs but couldn't remember if that makes steel harder or softer. So then I found a WHOLE SITE dedicated to DIY springs written in English no less.  I started reading it before deciding I was going off the deep end.

It's all there:

There are three basic principles in spring design:

    * The heavier the wire, the stronger the spring.
    * The smaller the coil, the stronger the spring.
    * The more active coils, the less load you will have to apply in order to get it to move a certain distance.

Based on these general principles, you now know what to do to change the properties of a spring you already have. For instance, if you want to make automotive valve springs a little stronger than stock, you can a) go to a slightly heavier wire and keep the dimensions and coil count the same, b) decrease the diameter of the spring, keeping the wire size and coil count the same, or c) decrease the number of active coils, keeping the wire size and spring diameter the same. Naturally, you can also go to a stronger material to achieve the same result.

So we simply need to do the exact opposite of the above. Simple... or not.

 If you got room to spare inside the barrel, you could try to increase the diameter of "working part" the spring. However, this may shorten the spring seeing how you have less wire to spare and may not leave enough room for the spring to buckle.

Tempering the springs may work - they should become softer when warmed. Their hardness mainly depends on the cooling rate. If you would swiftly submerge them in water when hot, they'd probably be stiffer and less elastic. On the other hand, slowly cooling them in some ceramic enclosure may give an elastic but really soft spring.

Probably test it on some spare springs you have around, e.g. from some ballpens you never use.


Anyway, don't forget to measure the required force on the feet first, a buckling spring with not enough force to make the contact is also useless.

[talis]

lol, thanks for the experiment, rippy!  80g, eh? hmmmm.
I'd stick mine under the coffee table but I only have one working M and I dont want to ruin it!  Someone who has some spare boards will have to try....

I'm pretty sure the time frame required to get any appreciable change in doing this would be rather long.

I plan to introduce Mr. Model M to Mr. Tablesaw this Friday, and get a jig together to test some of these ideas.

I think outside of changing the springs entirely, tempering may be the best approach, I'm just not sure how easy it will be to get a consistent result.

[wellington1869]

I'll try the propane torch trick and throwing them into a preheated oven to cool down.   If it work's I'll just fire up the Barbie (as the Aussies say) and do a couple hundred of them!

you could do a nice little side business roasting springs for people who want this mod

I plan to introduce Mr. Model M to Mr. Tablesaw this Friday

now that sounds like fun


I still say a stock part must be out there for this...

[talis]

The bandsaw won:

[talis]

Working on it.

[BucklingSpring]

How about adding weights into the keycaps? Should give an interesting feel!

That's exactly what I was thinking too. But it will change the inertia of the keys. Fast typists like some of you may find them too slow to return to position.

[BucklingSpring]

As my son loves to say, "Epic Fail".   Just as soon as you think you understand the Buckling Spring the thing gets all random on you.

I’m by no mean a spring expert but I know a little about them.
You need to be careful when you play with spring limits:

Elastic limit: Maximum stress to which a material may be subjected without permanent set.

Endurance Limit: Maximum stress at which any given material will operate indefinitely without failure for a given minimum stress.

Permanent set: A material that is deflected so far that its elastic properties have been exceeded and it does not return to its original condition upon release of load is said to have taken a "permanent set."

Unless you are a metal alloy engineer, you will have a hard time to predict the effect of stretching the spring to a permanent set. Heating the spring is not a good idea either. It will destroy the tempering and its elasticity.

Anyone knows a good old clockmaker? These guys have connections for small springs.

And I’m sure the folks at Unicomp will be welling to share the buckling spring specs.

[ch_123]

There are three basic principles in spring design:

* The heavier the wire, the stronger the spring.
* The smaller the coil, the stronger the spring.
* The more active coils, the less load you will have to apply in order to get it to move a certain distance.

Yeah, there are going to be some problems here. If you make the spring of too lightweight a material, it's going to wear out too easily, you can only make the spring so long before you start having to super-size the barrels and then the key size, so it seems we need more coils on the springs, but it's quite a tight squeeze as it is...

So, any experts on Nanotechnology here? =P

[BucklingSpring]

Maybe my son and I will try samurai swords next.

With all the comprehensive work you've done already, you must be ranking up there with the best masters in the art of swordsmanship.

It just took centuries of trial and error after all. But back then, they didn't have all the sciences you have. You got an edge on them for sure.

ROFL...

Thank you for making my day again. I was in desperate need of a good laugh.

[wellington1869]

ripster, what a great experiment. The slow cool one seems so close! What would happen if you only heated it to, say, 300 degrees? or 200? would that keep the force around 55g? (and therefore a better buckling feel)?

[huha]

Some people claim they can feel the difference in two piece keys versus one - I couldn't in my limited testing (although I will get another M Space Saving with two piecers so it'll be easier to tell).

I could definitely feel a difference between my Boscom and my Endurapro; the former does feature two-piece keycaps, but I'm not exactly sure if this makes for all the differences in feel.

-huha

[ch_123]

I could definitely feel a difference between my Boscom and my Endurapro; the former does feature two-piece keycaps, but I'm not exactly sure if this makes for all the differences in feel.

-huha

To me, it feels like the two piece caps dampen or absorb some of the tactility of the key.

[wellington1869]

so ripster, if 50g is too light for bs, what do you think is an ideal 'light' bs? 55g?
would it be possible to produce 55g by adjusting the temperature of the 'roast'?

[wellington1869]

Anybody have any luck finding a substitute online?  Lots of springs out there.  You can use one of my pics to count windings.

can you tell me the dimensions of the spring? (I guess in mm?)