Bicycle Mechanics - Will 9spd derailleur work for 7spd indexed system?

I got a mint 7spd (105 rear hub[126mm] with HG cassette) given to me. It's a mid '80s Japanese bike, I believe. Someone spray painted it "anonymous blue", so the heritage is unknown. But, it's lovely lugged steel and in nearly perfect condition. The only problem with the bike is that the RD took a major hit somewhere along the line and the pivot points are not "solid". Can I just install a new 105 RD(9spd) and have it index properly?

Does the RD really care about the # of indexed speeds? How do you gauge/measure this capability/capacity? What's the pertinent metric/s?

Fire away ;-)

yes

Does the RD really care about the # of indexed speeds? How do you gauge/measure this capability/capacity? What's the pertinent metric/s?
The RD is essentially a free travelling parallelogram with an internal spring moving it in one direction and the cable (therefore the shifter) moving it in the other. The "indexing" is a function of how much cable travel occurs based upon the shifter's movement of it.

So, the shifter determines the travel between cogs, and therefore must be matched to the cassette. The RD is essentially along for the ride and only follows orders. :)

The only limitation I can think of would be trying to use a RD intended for a small number of cogs whose travel may be limited too much to travel over the width of a newer standard cassette. In other words, a 126mm RD on a 130mm cassette. I am not sure if that is, in fact, a practical limitation, but probably easily guarded against in any case - use a newer RD on an older cassette.

The RD is essentially a free travelling parallelogram with an internal spring moving it in one direction and the cable (therefore the shifter) moving it in the other. The "indexing" is a function of how much cable travel occurs based upon the shifter's movement of it.

So, the shifter determines the travel between cogs, and therefore must be matched to the cassette. The RD is essentially along for the ride and only follows orders. :)

Well actually,the RD throw ratio does have to match the shifter, in index systems. Then it can follow orders properly.That's why a sram esp RD is not compatible with simano but a sram mrx RD is.Also the reason for much campy shimano incompatibality.

Well actually, ...
I figured I would learn where my mistakes were from you sydney. :p

Well actually,the RD throw ratio does have to match the shifter, in index systems. Then it can follow orders properly.That's why a sram esp RD is not compatible with simano but a sram mrx RD is.Also the reason for much campy shimano incompatibality.

That's true but not to the point in this issue.

The rear derailer has to match the brand of shifter.

Within brands there are a couple of incompatibilities:

Pre-1997 Dura-Ace vs. all other Shimano
Sram ESP vs. Sram Shimano compatible


However, within a given system the number of speeds does not matter to the derailer. The O.P. can use any Shimano derailer on his or her bike.

Sheldon "Lets Not Make It Seem Harder Than It Is" Brown

+----------------------------------------------------+
| A little inaccuracy sometimes saves tons of |
| explanation. --H.H.Munro ("Saki")(1870-1916) |
+----------------------------------------------------+

That's true but not to the point in this issue.

The rear derailer has to match the brand of shifter.

Within brands there are a couple of incompatibilities:

Pre-1997 Dura-Ace vs. all other Shimano
Sram ESP vs. Sram Shimano compatible


However, within a given system the number of speeds does not matter to the derailer. The O.P. can use any Shimano derailer on his or her bike.

Sheldon "Lets Not Make It Seem Harder Than It Is" Brown

+----------------------------------------------------+
| A little inaccuracy sometimes saves tons of |
| explanation. --H.H.Munro ("Saki")(1870-1916) |
+----------------------------------------------------+
Shouldn't that be any SIS shimano RD in an index system? My commment wasn't directed to the O.P., as yes was the correct, short, simple answer. Long answers are often needed to clear up misconceptions. Thanks for the imput.'...... sydney "just tellin it like it is" no carpet smoke ;) :)

Shouldn't that be any SIS shimano RD in an index system? My commment wasn't directed to the O.P., as yes was the correct, short, simple answer. Long answers are often needed to clear up misconceptions. Thanks for the imput.'...... sydney "just tellin it like it is" no carpet smoke ;) :)

All Shimano rear derailers made since about 1987 are SIS. As to the index issue, this is _only_ an issue with index systems...friction basically works with everything.

(Yes, I know the old Benelux coil spring derailers from the '50s needed a lever with more travel...) ;-)

Sheldon "How Far Back Do We Need To Go?" Brown
+-------------------------------------------+
| Any smoothly functioning technology |
| will have the appearance of magic. |
| --Arthur C. Clarke |
+-------------------------------------------+

Thank you all for your replies. I see now that I did not mention the shifters I am planning to use. I have some 8spd Ultegra barcons. So, if I understand correctly, I am good to go but I'll have one dead click.

So, it was apparent to me that for some given amount of pull on the cable (in length) you get a corresponding side-to-side change (in width) on the derailleur parellelogram. Since both of these movements are distance measurements, they can be gauged and the actual ratio can be determined. As I now understand (thank you gmason!), this relationship is linear and is what you are calling "throw ratio".

Apparently different manufacturers use different throw ratios, but not always. And, some manufacturers may have different throw ratios within their own product line.

Ok, so last two questions:

Is there a straightforward way to make the measurements and determine throw ratio? (I am thinking that this is easier said than done).

Is there some reference where someone has already done the work and it's all recorded? or, do we all have to rely on our good friend Sheldon for the specifics?

All Shimano rear derailers made since about 1987 are SIS. As to the index issue, this is _only_ an issue with index systems...friction basically works with everything.

(Yes, I know the old Benelux coil spring derailers from the '50s needed a lever with more travel...) ;-)

Sheldon "How Far Back Do We Need To Go?" Brown
+-------------------------------------------+
| Any smoothly functioning technology |
| will have the appearance of magic. |
| --Arthur C. Clarke |
+-------------------------------------------+
Probably nit picking,but I was thinking 1985. sydney "got a box full of old pre index shimano RD that wouldn't work" the pack rat .

Thank you all for your replies. I see now that I did not mention the shifters I am planning to use. I have some 8spd Ultegra barcons. So, if I understand correctly, I am good to go but I'll have one dead click.

In Shimano setups, the spacing is slightly different between 7- and 8-speed cassettes, but generally it's close enough to let you use 8-speed shifters with a 7-speed system.

There won't be any extra click if the derailer's limit screws are adjusted properly.



So, it was apparent to me that for some given amount of pull on the cable (in length) you get a corresponding side-to-side change (in width) on the derailleur parellelogram. Since both of these movements are distance measurements, they can be gauged and the actual ratio can be determined. As I now understand (thank you gmason!), this relationship is linear and is what you are calling "throw ratio".

Apparently different manufacturers use different throw ratios, but not always. And, some manufacturers may have different throw ratios within their own product line.

Ok, so last two questions:

Is there a straightforward way to make the measurements and determine throw ratio? (I am thinking that this is easier said than done).

It's not that easy to measure, but there is generally no need to measure it in practice, just match the brand of derailer to the brand of shifter, and the number of shifter clicks to the number of sprockets on the cluster.



Is there some reference where someone has already done the work and it's all recorded? or, do we all have to rely on our good friend Sheldon for the specifics?

See my latest article: http://sheldonbrown.com/speeds

Sheldon "Numbers" Brown

+--------------------------------------------------------+
| As far as the laws of mathematics refer to reality, |
| they are not certain; and as far as they are certain, |
| they do not refer to reality. --Albert Einstein |
+--------------------------------------------------------+

Thank you all for your replies. I see now that I did not mention the shifters I am planning to use. I have some 8spd Ultegra barcons. So, if I understand correctly, I am good to go but I'll have one dead click.

So, it was apparent to me that for some given amount of pull on the cable (in length) you get a corresponding side-to-side change (in width) on the derailleur parellelogram. Since both of these movements are distance measurements, they can be gauged and the actual ratio can be determined. As I now understand (thank you gmason!), this relationship is linear and is what you are calling "throw ratio".

Apparently different manufacturers use different throw ratios, but not always. And, some manufacturers may have different throw ratios within their own product line.

Ok, so last two questions:

Is there a straightforward way to make the measurements and determine throw ratio? (I am thinking that this is easier said than done).

Is there some reference where someone has already done the work and it's all recorded? or, do we all have to rely on our good friend Sheldon for the specifics?The simple answer is that the 8 speed ultegra barcons are compatible with a '9 speed' 105 RD with respect to the mechanics involved. Shimano 7 speed cog spacing is 5mm and 8 is 4.8mm. That is close enough that what you propose to do usually works.I've done it. Others say is doesn't work well enough for them.YMMV.