Fumbling around with my DKW Hummel

Several years ago I purchased a 1957 DKW Hummel moped. Being too slow to ride safely in modern traffic, I really only intended it to be a display item, but I did want to actually run. When I bought it, it did not run as the fuel tap was clogged. I only got around to replacing the fuel tap corks in February 2022 and instantly the moped fired straight up and I took it for a spin around the neighborhood. I reported on this here: https://dkwautounionproject.blogspot.com/2022/02/my-1956-dkw-hummel.html 

But the fix did not last. After another short ride the engine just would not fire up. After much frustrated fiddling, including replacing the fuel tap corks, I put the moped aside. Why invest so much time in such an impractical vehicle? I have many other more important projects to work on.

The fundamental problem is that access to the fuel outlet and tap is extremely limited. It can only be done by feel alone.

However, the Hummel problem kept niggling away at me. I needed to find out why it stopped working, so six months ago I drained the fuel tank and replaced the original fuel tap with a new reproduction fuel tap. I had avoided buying a reproduction fuel tap for a long time as I have found the quality of modern fuel taps to be generally poor, in addition to being extremely expensive at over 100 Euro (with postage to Australia around 85 Euro). That's a ridiculous expense for a novelty vehicle, but I felt I needed to give it shot and so bit the bullet. When the tap arrived, it proved less than satisfactory. The sieve that sits atop the tap and inserts into the fuel tank was too wide to fit through the M-12 sized outlet. I tried to crush it a little to make it fit, but this did not work at all and the sieve wire ripped up its seam. The lip of the sieve also prevented the M-12 bolt fitting to the petrol tank outlet. I was able to fit the tap without the sieve, but squeezing the on/off tap through the frame was very difficult. After fitting, I took the moped for another test run and after a few coughs and splutters, the engine steadfastly refused to run again. Back to square one.

This is the modern reproduction fuel tap - four parts broke after fitting - the sieve, the bronze tube, the plastic mechanism linking to the on/off switch and the fuel nozzle that connects to the plastic fuel line. Quality workmanship at exorbitant prices!

After more frustrating fiddling, I decided to remove the new tap and refit the original tap. This was an enormous problem as the tap was extremely hard to reach and the on/off tap that protrudes from the frame could not be removed easily as it is held in place by a split pin that is almost impossible to access. The removal had to be done in phases as the whole mechanism would not come out as one piece. To my great frustration the tap broke in several places - the on/off switch snapped off and the bronze inlet tube that inserts into the fuel tank and theoretically sits above the layer of residue broke. Fortunately I was able to flush the broken tube out of the tank with a flush of acetone. This troubling exercise however revealed the ultimate cause of the problem. The interior of the fuel was badly rusted and encrusted with decades of accretion. I had expected this but was hoping against hope it was not so as the Hummel 115 doesn't have a separate fuel tank; it is an integral part of the frame and cannot be removed.

So I needed the clean the tank 'in-situ'. I experimented with several different solvents and acids. Several doses of cleaning vinegar helped wash through a lot of the grit, but did not really treat the tank interior. Several different solutions of different acids were tried, but the last soaking of a citric acid solution gave the best result. After a week of soaking, I drained the tank and let it dry for a week before washing the tank through with acetate. This dislodged a lot of accretion. I cycled through the wash several times until the rust colour finally disappeared.

So now I had a 'cleaned' tank and two broken fuel taps. I tried to reassemble the original tap using some spare parts from the reproduction tap. Of course, the reproduction tap is completely different sized to the original so that proved fruitless, but it gave me an idea - why not dispense with the original tap entirely and run a fuel line directly from the fuel tank through an inline filter with an inline tap? Great idea! The new tap had a fuel nozzle that could fit directly onto the fuel tank outlet, so I mounted that in place and attached a fuel line to the nozzle and.... the nozzle broke in half. What the hell? Another case of a brand new fitting being of such poor quality that it simply snapped in two when a rubber fuel line was pressed onto it.

I did not want to buy another fuel tap so I dug through my boxes to spare parts and managed to find some parts of an MZ fuel tap for my IWL Troll scooter. The nozzle for that tap roughly fitted into the fuel outlet so I was able to jury rig a new fuel line and plug it into the carburetor. It was by no means pretty, but as soon as I turned it on and started pedaling, the moped fired right up again and I was away. I'm quite amazed at how well the little engine runs despite being unused for several years. I managed to take on a test ride around the neighborhood twice....

Yes, you have to unlock the side panel to switch on / off the tap, but at least it works.

That all said, the set up is very much temporary. The MZ tap fittings don't fit correctly so petrol leaks all over the place, but the fix generally works. I'll try and fix the leak and make the set up more robust. I know, it's ridiculous, but.... It is satisfying to hear it run again.

1953 IFA F9 Project - Buchan Motorsport

Back in 2016 I began to consider purchasing an IFA F9 or F8. I made several enquiries with oldtimer dealers in Eastern Germany. One of those was KFZ Steffen Lange of Calau. Steffen Lange had a large collection of DKW, IFA and assorted East German vehicles for sale and I often dreamed of loading a shipping container of 'goodies.' One of the vehicles I seriously considered purchasing was this 1953 IFA F9. This is a very early, IFA-built F9s constructed in the former Audiwerkes in Chemnitz. Only 1627 examples of the limousine were ever built as IFA struggled to overcome material shortages, build quality issues, lack of workforce and competing priorities with VEB's demand for truck and tractor production. For the details of these challenges, see here:
https://dkwautounionproject.blogspot.com/2020/11/postwar-veb-horch-trucks-and-tractors.html

Among the 'primitive' and distinctive features of these early cars are the split front and rear windscreen, trafficators and a front mounted petrol tank, as East Germany did not have a fuel pump manufacturer at this time.

The car had been sitting for decades and required extensive rehabilitation. From an initial glance, I suspect there have been substantial modifications made to the car's dashboard and gear shift. The fascia and column mounted gear shift appear to come from a later, Eisenach-built F9, however, it could be that this was an early Eisenach assembled car using an IFA-built older body. A transitional example.

After due consideration, I decided not to proceed and ended up buying a 1953 DKW F89P instead. The long restoration process of that car is extensively documented in this blog.

I'm pleased to see however, that the owner of Buchan Motorsport has purchased this rare and interesting car. They have produced two every well-made and detailed videos about the F9 and its importance and this car makes a great, practical addition to their interest. I look forward to seeing more videos about the progress of the car's rennovation.

Part One - DKW F9 - The Forgotten Superstar

Part Two - DKW F9 - How it influences German car industry to today

Part Three - I Bought my own F9

Part Four - First Fixes

Part Five - How much F9 is in a VW Phaeton?

Part Six - Updates (rear lights and trunk)

Part Seven - Back in Black


Part Eight - New Look


Part Nine - Exterior Finished


The Buchan Motorsport documentaries:
https://dkwautounionproject.blogspot.com/2022/12/dkw-f9-forgotten-superstar.html
The history of the Development of the DKW F9:
https://dkwautounionproject.blogspot.com/2017/07/1939-dkw-f9-prototype.html
The history of the Development of the IFA F9:
https://dkwautounionproject.blogspot.com/2020/07/the-development-of-ifa-f9.html
The history of the Development of the DKW F89P:
https://dkwautounionproject.blogspot.com/2017/06/dkw-f89p-new-meisterklasse.html
Winfred Kuhl's IFA F9 prototype project:
https://dkwautounionproject.blogspot.com/2021/11/1948-ifa-f9-project-part-four-body.html

DKW F9 - The Forgotten Superstar - Buchan Motorsport

I would like to give a big shout out to the team at the Buchan B-Sport YouTube channel for putting together two really excellent videos about the history of the DKW F9 and its modern successors. These are really well put together and very interesting. Please enjoy and give them a like. www.youtube.com/@bsport320




DKW F9 development:
https://dkwautounionproject.blogspot.com/2017/07/1939-dkw-f9-prototype.html
DKW F89P development:
https://dkwautounionproject.blogspot.com/2017/06/dkw-f89p-new-meisterklasse.html
IFA F9 development:
 https://dkwautounionproject.blogspot.com/2020/07/the-development-of-ifa-f9.html

The Buchan Sport channel owner has now purchased an IFA F9 for restoration. Interestingly, I had looked at this car several years ago with an eye to buying it. I didn't proceed however as I thought it might be too much of a challenge (little did I know!). https://dkwautounionproject.blogspot.com/2023/01/1953-ifa-f9-project-buchan-motorsport.html

1931 DKW R1000 Stationary Engine

In 1919, Jorge Rasmussen and Hugo Ruppe began showcasing an 18cc 'toy' stationary motor nicknamed "the boy's wish" at industrial shows. Often described as a toy motor, there was scarcely a market for such an expensive frivolity in post-Great War Germany. The true purpose of this first 'DKW' was as a demonstrator stationary motor targeting the small business market.

In the years before the Second World War, DKW would become the world's biggest motorcycle company as well as a major German automobile manufacturer, but a gigantic part of the company's business was stationary and industrial engines. These ranged from small, air-cooled single cylinder two-stroke engines to eight cylinder four-stroke engines, air-cooled and water-cooled, for use in cars, tractors, mowers, draisons, boats, water-pumps, fire-hoses, sprayers, electrical generators, in two-stroke, four-stroke and diesel format.


The stationary motor division of DKW was actually at the forefront of engine development, trialing every possible format. In 1931 they presented a four-cylinder inline water-cooled two-stroke motor, designated the RVW. The engine came in 1000cc and 1200cc capacity, with the 1200cc version generating 29 PS. Sales however proved to be very disappointing and only 100 engines were built, serial numbers 299001 to 299100. The engine was still being advertised in the 1933 brochure above.

Given the small production numbers, survivors are few. A complete example is held in a German museum collection (see postcard at the top of the post). A second example was discovered by Roland Servais in a Belgian scrap yard. Serial number 299035 was largely intact but some parts were missing.




Roland restored the motor and mounted it on a steel chassis and connected it to a French dynamo built by "Ateliers de Constructions Electriques de Lyon et Dauphiné" around 1925. You can see the engine in action in the video below:

Roland has decided it's time to pass the engine to a new custodian, so if you are interested in purchasing this extremely rare piece of DKW history, you can contact him here: rolandser@gmail.com
The engine is located in Belgium. Roland has other stationary motors too.

So why did DKW abandon development of the four-cylinder two-stoke? This is an interesting question, given that in a few years time DKW would come to recognize that the 600cc and 700cc twin-cylinder had reached its developmental limits and would begin scrambling to develop a new engine for their automotive range. They would, of course, settle on the 900cc triple, which in its initial configuration generated approximately 26 PS. In the RVW engine they had a motor that could generate at least 29 PS. It cannot be that this format engine wasn't suitable for an automobile engine because a remarkably similar engine was powering a range of Czechoslovak cars at roughly the same time - and this is where the story becomes intriguing.

Zbrojovka Brno was a Czech armaments manufacturer that branched out into automobiles in the mid-1920s. After developing their own series of two-stroke budget cars, in 1932 they sought to enter a licensing partnership with DKW. Two engineers were dispatched to Zwickau to work on the project, but DKW's licensing fees were too high for Zbrojovka and they walked away from the deals. The engineers however had learned enough from their time in Zwickau to design their own car from scratch in a period of only 9 months. This would be the Zbrojovka Z-4 ("Strong Four"), which went on sale in January 1933. The Z-4 was powered by a four-cylinder water-cooled two-stroke motor of 908cc, generating 16 PS. Later series were upgraded to 980cc capacity. The engine is strikingly similar to the DKW RVW, so it seems entirely possible to me that Zbrojovka's engineers had inspected the RVW engine in detail, or purchased one and reverse engineered it. Of course, this is only speculation but it's a plausible theory. Zbrojovka also used twin-cylinder two-stroke engines, derived from DKW technology.

Zbrojovka weren't the only Czech concern to seek a licensing agreement with DKW at this time. Jawa would enter into a licensed arrangement to manufacture the DKW F2 in 1934. DKW two-stroke motor technology would go on to power both their automobile and motorcycle range well into the 1990s.

Jawa car history: https://dkwautounionproject.blogspot.com/2021/02/jawa-700.html 
Zbrojovka Z-6 brochure: https://dkwautounionproject.blogspot.com/2019/10/1935-zbrojovka-z-6.html

1938 DKW SB350 Restoration Update - Coal Mining

After taking the bike for quick test ride it was apparent the engine was low on power, especially in first gear. There were a couple of items I decided I needed to check. The first was to clean out the exhausts, which I was certain were carrying some 80 years of baked in oil residue. The exhausts came out easily enough and quick tap confirmed my suspicions.

The lumps that simply fell out of the exhaust pipes were like little lumps of coal!

After a good tapping I had more than a handfull of coal.



A view of the baffles at the fishtail end. Pretty black.

Next step was soaking the exhaust parts in kerosene to loosen the oil residue.

A look down the silencer shows the state of the baffles. After soaking in kerosene I was able to scrub out a lot of the residue. It wasn't perfect but it was enough to clear the baffles.

Someone suggested to me that you could burn out the remaining residue, so I gave that a go. Unsurprisingly the exhaust 'fired up' and burned for about half an hour, relighting itself periodically with a big 'whoop!' After another round of cleaning up and scrubbing off dirt, mud, oil and rust, I reassembled the pipes and refitted them.

I took the bike on a test ride to the Arthur Grady Day on Saturday 17 May 2025. The exhausts smoked a bit and needed some tightening up afterwards. Power is still low, especially in first gear. Next thing to do will be to re-look at the gearbox and clutch lubricant as I think it may not be right. 
https://heinkelscooter.blogspot.com/2025/05/arthur-grady-day-may-2025.html

Previous step: https://dkwautounionproject.blogspot.com/2025/05/1938-dkw-sb350-restoration-may-update.html

1938 DKW SB350 Restoration May Update 2025

I picked up my DKW SB350 at the end of April but I did not get a chance to do anything with until the 1st day of May.

First things first was to try and start it up myself. The choke needs to be on (lever turns towards the rider) and after only a few kicks - during which is made almost no compression noise at all - it fired straight up. The engine quickly settled into a steady idle. Even my 1954 DKW RT250 doesn't idle reliably unless its had time to warm up. After a quick walk around to test everything, it was time to take it for a test ride.

Wayne and Kim had both test ridden the bike as they worked on it and neither had been particularly excited about it. Both reported it was complicated to ride. There were good reasons for this. When Wayne took it for its first test ride, the clutch was barely functional. The clutch has now been recorked. Kim took it for a test ride after reinstalling the clutch, but he found that the replacement clutch cable was about a foot too short, so used the foot clutch. Using the left foot clutch and the right hand gear shifter requires quite a bit of practice on an unfamiliar bike. So I had been prewarned.

I decided to reuse the old clutch cable which, being original, had decent tension adjustors at both ends, so I was able to install and get the correct tension on the hand clutch lever. I must say, I have been very frustrated with many of the reproduction parts, especially the cables. They never fit and are often the wrong length - in this case a foot too short! That's completely unacceptable. This meant not having to worry about the foot clutch pedal. I quickly realised that the foot gear shifter was difficult to use due to its positioning, but the hand gear shifter worked well, so I stuck with that.

My first riding impression was of great instability. This was due to the seat which was not firmly mounted at the front bracket, causing the seat to slip several millimeters each way as you turn. This movement was very disconcerting but was an easy enough fix. I removed the (non-original) bolt securing the seat to the frame and packed in several washers to stop the front of the seat slipping. This allows for the rear-mounted springs to provide cushioning, but not slip from side to side.

The rear wheel foot brake (right side) feels poorly placed. Its obviously placed where it is for heel use, but it needed a lot of adjustment to stop the bike effectively without needing to move your foot too far back. Adjustment was easy enough as the brake rod has a decent adjustor. Fortunately the hand brake was extremely effective and I didn't need to use both brakes on the short, rather slow test ride around the block.

Which brings me to the last thing - power. The engine ran really flat and it was hard to accelerate. There are two things to look at: 1. the exhausts need a damned good decoking. At the silencer end they are completely filled with oil. After another test ride to confirm the brakes and seat issue is solved, I'll remove the exhausts and give them a good clean. And 2. I need to check the carburettor settings and possibly the jet size. By all accounts these bikes will comforably do 110 kph. I think I'd push it to get over 50 kph at the moment.

All in all, a good first test and I think I know what I'm doing now. A big thanks to Wayne, Kim and Gavin for helping get this far.



Next step: https://dkwautounionproject.blogspot.com/2025/05/1938-dkw-sb350-restoration-update-coal.html

1938 DKW SB350 Restoration Update April 2025

After we discovered that the clutch was not really operable, the clutch basket was pulled from the bike and I ordered a new set of cork plugs. These arrived surprisingly quickly from Germany on 11 April. See the proceeding post for the details here: https://dkwautounionproject.blogspot.com/2025/04/1938-dkw-sb350-restoration-update-march.html

I had cleaned up and polished the metal plates in the proceeding weeks. The condition was pretty good as far as I was concerned.

I had some time over Easter so I set to it. The cork plugs are twice the height required in the clutch, so each plug was cut in half using a craft knife. Some plugs were a little more accurate than others but all will need to be sanded down to a flat surface when they're mounted in the plates.

The cork needs to be boiled to make it soft and pliable enough to fit in the holes. Half an hour on a rolling boil was enough. It must be noted that once out of the hot water they cool down very quickly, so it's best to fit them in small batches while the rest 'cook.'

To my surprise, the cork was very easy to install. There was some inevitable losses due to seams and weaknesses in some of the cork plugs, but the bags contain more than sufficient for two sets of clutches.

The bottom plate was actually the easiest to fit.

It took about half an hour to fit all the small plugs. The big plugs follow the exact same process.

Almost done.

Job completed. It was a very easy and quite satisfying task. Now, the cork needs to dry as they were thoroughly cooked out.

Once it was dry I gave each of the cork plates a good sand to make them smooth and flat. The cork sanded very easily.

The clutch 'reassembled'. The clutch plates will need a finer sanding to get them to the right thickness and fit. I will leave that to the experts to do as I don't want to sand anything down too far that would undo the good work above. Hopefully this will be the last fix before the bike can be ridden.

And I received the call a few hours after I dropped off the clutch that the bike is now running, rideable and ready for collection!

I've got a few little things to sort out and then learn how to ride it.

Restoration update May 2025: https://dkwautounionproject.blogspot.com/2025/05/1938-dkw-sb350-restoration-may-update.html

1938 DKW SB350 Restoration Update March 2025

It's been a while since I've reported on progress with the DKW SB motorcycle so a recap is in order. I dropped my bike to some mechanic friends to sort out the carburetor and ignition issues back in November 2024. Those issues were quickly resolved, but like the proverbial loose thread on a jumper, other issues quickly unraveled. https://dkwautounionproject.blogspot.com/2024/11/1938-dkw-sb350-restoration-update.html

New tyres were obtained but when the guys removed the old tyres they found that the rims were completely rusted through from the inside. It was probably the rust that was holding them together. New rims were built from NOS stock rims and new spokes.

As the rims were new, they needed to be painted to look 'age appropriate.' They look great.

The guys managed to remove the old speedometer from its housing and install the new speedometer. Apparently there were some very tiny screws in the speedometer back holding it on. The original speedometer housing looks much better than the modern reproduction, which was not a match at all.

Next step was to start her up and take her for a test ride. The clutch proved to be barely functional. Well, after 80 years this is hardly a surprise. We pulled off the covers and took a look at the clutch.

Prewar DKW motorcycles used plates with cork plugs, approximately the same size as a cork bottle stop. Over time these decay and fall apart. As soon as we opened it up some cork plugs fell out.

This ring of cork plugs looks like it was original. All the plugs were consistently sized but blackened and inflexible. You can see some plugs are missing from the disc.

This ring of cork has been repaired / replaced at some point. Some cork plugs are original but replacements have been cut from square pieces of cork, probably from a cork bottle.

It's much more apparent in this disc. A little pressure on the plugs and they all came out easily enough. The metal plates were in good condition.

A new set of corks is on order. Below is a video showing how the cork is fitted.

This will be my next job!

After soaking the plates in petrol....

The cork arrived on 11 April 2025. It only took two weeks from Germany. Generally post from Germany takes 4-6 weeks, but it was a light package.

November 2024 update: https://dkwautounionproject.blogspot.com/2024/11/1938-dkw-sb350-restoration-update.html

The next step - April 2025: https://dkwautounionproject.blogspot.com/2025/04/1938-dkw-sb350-restoration-update-april.html