SP (Shutter precision) small hub dynamo PD-8 & SD-8

Tested from: 12 Agust 2011

SP has introduced their newest dynamo hub at the Taipei cycle show, which I've been waiting for for a while (since late 2010). It doesn't have a switchable dynamo but it's still very cool. Its specifications are better than the SONdelux in efficiency, yet its power output is higher when needed (for multiple LEDs or devices with USB power output), it is StVZO compliant on its own in 622mm wheels which the SONdelux isn't. Mass is about the same as the SONdelux. The only thing the SONdelux has that the 8 series hasn't is the pressure equalising system, but the price of the SP is about half that of the SONdelux.

If this dynamo runs as well as SP's switchable hubs with which I can hardly feel any vibrations on my test bike with Edelux headlamp, then it will be a great hub. But, that's yet to be tested...


Mass: PD-8: 407g (disc brake version, 32 holes, without a quick release axle, 3.0W), SD-8: 390g (disc brake version, 32 holes, without a quick release axle, 1.8W/small wheel version). The rim brake versions are lighter see the specification table.
Construction: 26 poles. No earth on the axis, excellent! Bearings are sealed cartridges.
User serviceable: No, unless you make your own tools (and worse, the axle ends are round since I think 2013... I will see how to deal with that). If the bearings wear out after the warranty, that can be done for a cost of $30, there is a service centre in Germany (possibly others, I will see).
Price: currently ca. € 150,- / $180,- (note that it's hard to get hold of at the moment, prices have risen since early 2011 for a few reasons, one being that magnet prices have increased quite a bit since spring 2011).
Dimensions for spoke calculation: For Sapim spoke calculation page: K=100 mm, Ka=25 mm, Kb=25 mm, D1 gear=52 mm, D1 non-gear=58 mm (it's a bit annoying that the flange diameter is different for the 2 sides of the disc brake hub (the SV-8/PV-8 rim brake hubs are fully symmetrical, K=100 mm, Ka=25 mm, Kb=25 mm, D1 gear=52 mm, D1 non-gear=52 mm), otherwise wheel building would be trivial).
[ K=hub width, Ka=distance of flange to hub-end, Kb=same, D1 gear=diameter of circle through the spoke holes on the gear side, D1 non-gear=diameter of circle through the spoke holes on the non-gear side ]

Pdf from SP with specififcations and efficiency information:


Note the small opening next to the axle in the last picture.

It's small!:

Built into a wheel:

Available types of the series 8 dynamo hubs

Specifications (100mm hub width, the small wheel versions are also available in 74mm hub width):

  PV-8: smallest/6V3W   for 622mm=700C/rim brake   390g
  SV-8: smallest/6V2.4W for 622mm=700C/rim brake   367g   (=6V3W for 20 inch)

  PD-8: smallest/6V3W   for 622mm=700C/disc brake  410g
SD-8: smallest/6V2.4W for 622mm=700C/disc brake  385g   (=6V3W for 20 inch)
(Actually, small-wheeled dynamos are more like 1.8W dynamos when spoked into a 559mm or 622mm rim)

Experiences with the PD-8 (3W dynamo hub for large wheels)

The dynamo axle is from aluminium. Some people have broken the aluminium axle in the Shimano DH-3N80 and for heavy use the DH-S501/DN-3N72 is recommended instead on some forums, so I'm not sure if the weight savings are worth it but see my discussion about tradeoffs at the end of the page for more on this. A problem I had was simply getting the skewer through the hub's axle. It's in at least 2 parts and the skewer's axle was hard to get through the PD-8, and it got stuck in the SD-8 from whichever side I tried. I pushed it through with force and a tiny piece of aluminium from the inside was taken off by the threading on the skewer axle. This is not nice even though it won't (or rather: shouldn't) affect durability.

When I noticed these dynamos only have 26 poles like Schmidt's dynamos, I immediately wondered whether that would mean more noticeable vibrations in the handlebar (compared to SP's series 6).

2011-8-12/13 (at night): First ride: I notice the vibrations, barely. As with the PD-7 (HB015) there's not a sudden increase in vibration that I feel with all other dynamo hubs at around 19-22 km/h. This could be just because the vibrations are less than other dynamo hubs. I do think the vibrations are slightly stronger than with the PD-7 (HB015), and as there are now 26 instead of 28 poles, this doesn't surprise me. I will try again with the same bike using the PD-7 (HB015) and the Shimano DH-3N80 to confirm my experiences (again).

2nd ride (on very good asphalt, butterfly handlebar with soft foam rubber): Hard to feel the vibrations.

3rd ride with DH-3N80 (on very good asphalt, butterfly handlebar with soft foam rubber): Easy to feel the vibrations, also the start (at about 22 km/h, ends at about 28 km/h) is again more obvious.

Next (this weekend) I will use a different handlebar (flat bar), the one I originally used until recently on that bike, as that may also influence the results...

2011-8-14: 4th & 5th ride, 2 x 10km on various types of asphalt (flat handlebar with San Marco & Brooks leather grips, both harder than the foam rubber on the butterfly bar): I can feel the start of vibrations at 22 km/h and it ends at about 28 km/h. The handlebar makes a noticeable difference, but the vibrations of the PD-8 with a flat handlebar are also not annoying (just).

Ride with the PD-7 (HB015) on the same bike/flat handlebar: Here the start of vibrations is almost impossible to pinpoint, the end is clear though, at about 26 km/h. This confirms that the reduction of 28 to 26 poles in the SP series 8 causes stronger vibrations at slightly higher speeds, which is of course as one would expect. This means all in all for me the PD-7 (HB015) will almost certainly remain my favourite hubdynamo...

Ride with the Shimano DH-3N80 on the same bike/flat handlebar on various types of asphalt: Start of vibrations at about 21 km/h, end of vibrations at about 26 km/h. Compared to the rides just before this one it is apparent again that the vibrations with the DH-3N80 are stronger.

2011-8-15: Ride of 2x9 km with the Sanyo NH-H27 to confirm the feelings on the test bike despite some changes (headset was tightened a bit for example at the end of 2010 after I had done all tests with the NH-H27): The tighter headset means it's a harder to feel the start of vibrations at about 19 km/h, the end is clear at about 25-26 km/h. The vibrations seem a bit stronger than of the SP PD-8. So my feelings of which dynamos are best regarding vibrations hasn't changed after this test.

My preference in dynamo hubs regarding vibrations is:

  1. SP SD-8 (can't feel any vibrations, however, power output is noticeably lower than with 3.0W dynamos
  2. SP PD-7 (HB015) (least amount of vibrations, still need to check high speed up to 50 km/h)
  3. SP PD-8 (worse than PD-7/HB015, better than Sanyo/Shimano because the start of vibrations is hard to feel and they are slightly less strong than with the Sanyo/Shimano hubs, still need to check high speed up to 50 km/h)
  4. Sanyo NH-H27 (vibrations start sooner and end sooner than the DH3N80, namely 19-25 km/h, high speed test still to do)
  5. Shimano DH-3N80 (better than the SON28 at all speeds up to about 43 km/h, at that point the DH-3N80 has a very strong vibration much worse than the vibration of the SON28, however, for most people this speed is uncommon).
  6. SON28 (this is the only hub with which I could even feel vibrations with light off on my standard test bike!).

Tests with the 3x XM-L: To come.

More information/experiences/pictures to follow...

Conclusion for the PD-8 (preliminary, until triple XM-L and other headlamp testing is done):

Appearance: Looks very nice, I like the look of all the SP hubs actually.

Vibrations are slightly worse than the best one in this respect that I've tested so far, which is the SP switchable hub PD-7 (=HB015).

If you want the lowest weight dynamo hubs the SP hubs are the ones to get, because they even beat the SONdelux in that respect. The 3W rim brake version PV-8 is the same weight as the SONdelux (which is NOT a 3W dynamo hub with respect to the StVZO regulations, as it's originally meant for small wheels). This weight comes with some tradeoffs and one of those is the increase in vibrations of the PD-8 compared to the PD-7 (HB015) by reducing the number of poles to 26, the same number as the SONdelux. SP wants to target those people who want the lowest weight/highest efficiency, and to be honest, I'm not one of those people, I go for other things. For me vibrations count first, therefore my favourite dynamo hub remains the PD-7 (HB015)...

Note that vibrations in the handlebar differ per bicycle, some people never notice them. That's because of the eigen-frequency of the fork+ wheel being such that they don't get a strong enough vibration, because this is caused by resonance.

Still to do: High speed testing at 40-50 km/h to see what happens with vibrations (the Shimano dynamo hubs have a very strong high speed vibration which is bad, so I hope the SP doesn't have that). Running a triple XM-L (should work just fine), Running a Philips SLD to see what happens with vibrations.

Experiences with the SD-8 (3W dynamo hub for small wheels; when used in a large wheel it is a bit less than 2.4W)

I have spoked in the SD-8 (the small wheel/2.4W dynamo) and I will soon see what happens with vibrations with an Edelux, but especially what happens trying to run multiple LEDs from this dynamo, in particular the Philips LBL with the special dynamo driver, and a triple XM-L with the special dynamo driver.

First rides (2011-9-26 at night ca. 02:00)

With Edelux: The light comes on quite late. I rode on the path from my house to the road fairly slowly, ca. 5km/h, and the Edelux did not come on. I checked to see if something had come loose! But no, it worked when giving the wheel a strong push. I could not feel vibrations at any speed I tried. I tried switching the Edelux on/off while riding, and still didn't feel anything.

This seems to confirm the idea C.Tsai had that the reason the Sanyo hub is less noticeable is that it is not really 3W using StVZO setup (when testing for various dynamo drivers the output of the Sanyo NH-H27 was not as high as with other dynamos...). However, the Sanyo is not vibration free, the SD-8 is. On the other hand at low speed the Sanyo gives a little more power as it doesn't have the low speed problem of the SD-8. I suppose I should now try out a SONdelux and see how it compares! As I strongly prefer 3W dynamos I'm not sure it's really worth doing though.

With the special dynamo driver running a triple XM-L: I had the same problem of the light not coming on for a while. I didn't feel vibrations. I suppose the chopping up combined with 'too little vibration to cause noticeable resonance' means at higher power output vibrations with this driver are also not noticeable when using it with this dynamo hub. I will check what happens with the PD-7 (HB015) and PD-8 the next few days.

I felt full power with the special dynamo driver running a triple XM-L with the Shimano HB-NX70 came at about 30 km/h, I think this is what the developer said as well. With the PD-8 and PD-7 (HB015) I still need to test this but I will do so tomorrow night. With the SD-8 I felt that brightness kept increasing at least until 36 km/h. This is to be expected I suppose.

The results of these tests with the SD-8 made me think back to the PD-8 and I now feel even stronger than before, that if SP had made the PD-8 with 28 or 30 poles it could have been near perfect: vibration free and delivering enough power at all speeds for lighting and USB devices etc. which is something that people will use them for fairly soon...

2011-9-27 at night ca. 01:00

Using Philips SLD, at low speed (walking pace or even slightly less) light comes on, goes off, etc. This doesn't happen with a 3W dynamo this way.

Preliminary conclusion for the SD-8, used in a wheel with 622mm rim:

Comments on design and tradeoffs

Note: I discuss the PD-8 but also the small wheel version SD-8 in this section as the competition at the moment for the SP series 8 hubs (in mass and efficiency) comes only from the SONdelux (also a small wheel dynamo). I will try to separate it all more to eliminate any confusion (between small and big wheel versions etc.)

The PD-8 and PV-8 are the lightest 3W dynamo hubs, and the most efficient ones. The small wheel versions SD-8 and SV-8 are even lighter (beating the SONdelux) but have, when used in a big wheel (559mm or 622mm rim) less power output than a big-wheel dynamo hub such as the PD-8/PV-8/SON28/DH-3N80 etc., which means less light at low speed, or less power for running multiple LEDs. This is of course the same limitation that one has when using a SONdelux in a wheel with 559mm/622mm rim. But many people have been using the SONdelux (which is really nothing more than a renamed SON20R) and before that the SON20R in big wheels which is also why I'm going to test the SD-8, and see how much one notices of this lower power output of a small wheel dynamo in a big wheel.

I feel SP went a bit too far trying to improve on weight to get the small wheel version to about the same weight as the SONdelux for the disc brake version (the small wheel version SD-8 is 390 g, the SONdelux is also a small wheel dynamo and weighs 390g for centre lock according to Schmidt), and lower than the SONdelux in case of the rim brake version, which is 367g according to SP, whereas the SONdelux is 390 g according to Schmidt. Some of the weight loss can be attributed to using aluminium hub ends by SP, whereas Schmidt uses stainless steel.

Of course I cannot compare the PD-8 to a Schmidt hub, as Schmidt hasn't released a SONdelux type 3W hub. Not yet anyway, they will do so end of August 2011, but mass will be higher than the SP. September 2011: Specifications are now up on Schmidt's website: 440 g. for the SON28-new rim brake, whereas the PV-8 is 390 g., so the SP rim brake version has an advantage of 50 g. on Schmidt. The SON28-new for 6 bolt disc brake is 460 g., so the SP has an advantage of again 50 g. Schmidt also seems to make the 6 bolt IS disc brake hub with dissimilar flange diameter left/right...

All the best and low weight dynamo hubs (Shimano DH-3N80, SONdelux, SP hybrid series, SP series 8) use aluminium axles. There have been reports of broken axles with the DH-3N80 and for heavy riders/loaded touring the DH-3N72 is then perhaps better. I have no data on things breaking with my hubs, they all still work :)

For SP, the hub ends are aluminium, perhaps no problem at all, but aluminium is soft, so it makes me a little uncomfortable. What I'm thinking of is galvanic corrosion between the hub ends and a steel fork eating away at the aluminium (and then losing grip which is essential for a dynamo hub to prevent the axle rotating). Perhaps not a problem at all?

I can understand that SP want to beat the SONdelux (in weight/mass), but I feel some opportunities for a near perfect dynamo hub have been lost. With the switchable hub (HB011, HB015, now called PV-7 and PD-7) SP was very close to a near perfectly running hub, and one that is fully symmetrical which makes wheel building easy. I would liked to have seen the same symmetry and 28 poles in the newer series 8 which would have given an almost perfect dynamo hub. Now the flange diameter is not the same left/right for the disc brake hub, which is unlike the prototype specifications (but of course it is now lighter than the prototype specifications). It has 26 poles instead of 28 which makes the vibrations noticeably stronger, so I would have liked to see 28 poles, or even 30. I don't know what the manufacturing problems are for such small dynamo hubs, perhaps more poles is very hard? But, the less poles, the stronger each jolt and where the series 6 (HB011, HB015, now called PV-7 and PD-7) switchable hub was nearly vibration free, the series 8 is much more noticeable in that respect. This means for me, as I'm not a weight-weenie, and as I ride a lot on roads with high quality asphalt where I feel those vibrations, that I prefer the SP PD-7 (HB015) dynamo hub.

Dissecting a PD-8 (version from 2011, with partially flat hub-ends for a spanner)

2016-2-28: I started long ago to take a PD-8 apart, actually in 2012 I think it was... Recently I finalised it, made pictures, and wrote this section. The hub ends of the original versions have 2 flat edges on the ends and can be removed with a spanner, newer ones would need some sort of tool with rubber that grips them enough to unscrew them if they are screwed on, perhaps they are now friction-fit like the PD-8x (on the original PD-8 there is locktite or similar on the threads). But how to continue? First I screwed it all back together but found that the hub didn't output any power any more. This meant I didn't feel the need to be careful any more :)

I returned to dissecting it. I let it lie in the oven for half an hour at ca. 110 degrees to loosen any adhesive that I was sure had to be used to keep the 2 sides of the hub shell together. When unscrewing the 2 hub ends I found that 1 hub end didn't want to come off. I overtightened it I think when putting it back on the previous time... So it took some effort with brute force. The cap next to the dynamo's power output then had to be removed, again with brute force, I think by screwing together the hub too tightly it got twisted on the axle, or something (it wasn't centred any more...). Anyway, the dynamo itself is very tight in the aluminium shell and the hub is clearly made with such a design and small tolerances, that taking it apart yourself to remove the bearings, is not an option. But there is nothing that requires this! This is in my view a design flaw, as I said long ago, and seeing the hub open, is interesting but it's not cool that it's so hard to change bearings. They still need to be pressed out of the aluminium shell if you want to replace them. I pulled the dynamo out of the 2nd half of the hub's shell and you can see this on the pictures.

Next to come, dissection of a PD-8x, which got defective from lying sideways in the rain (so water could seep in for too long this winter... This is a general problem and can/will happen with just about any hub that way)

pd8-gedemonteerd-1 pd8-gedemonteerd-2 pd8-gedemonteerd-3 pd8-gedemonteerd-4 pd8-gedemonteerd-5

To email me go to the email page