The John Deere 420 Crawler I Bought in a Gravel Lot: Three Mistakes, Roughly $2,300, and the Checklist I Actually Use
September 20, 2023. Gravel lot behind a farm auction in central Illinois. I'm standing there with a clipboard in one hand and a cashier's check for $4,200 in the other, staring at a 1956 John Deere 420 crawler that turns over but won't catch.
I've been handling parts and equipment purchases for a nine-crew landscaping and site-work outfit for seven years. I've made a lot of errors in that time. Three years ago I paid for a hydraulic rebuild kit before confirming the serial number break, and we ate $890. In my first year (2017) I ordered frost-proof spigots instead of the frost-free version and had to explain the difference to two very annoyed crews. I keep a running list. That list is at 23 entries and somewhere north of $19,000 in documented waste.
The 420 crawler added three more entries in eleven weeks.
The plan, which sounded reasonable in September
We needed a small dozer to cut and level a couple of residential lots. A new or late-model unit runs well past $40,000, and honestly we didn't have the work volume locked in to justify it. A 1950s crawler made sense on paper: cheap to buy, mechanically simple, parts still floating around. Ours had the two-cylinder gasoline engine — the 420 was offered with a few engine options, so I checked, but the two-cylinder gas unit was what showed up.
My estimate to the owner was one week to running condition. He wrote it down. That detail matters later.
The crawler itself wasn't the expensive part. The expensive parts were everything I decided to build, buy, and bolt around it — which is where the next two months went.
The breaker box decision I made with my gut instead of a voltage table
Before the crawler could move, I needed power at the equipment shelter for a stick welder, a compressor, and a battery charger. That meant a breaker box and a couple of circuits.
Here's the honest version of the decision: I went back and forth between hiring an electrician for the whole job and doing the enclosure and circuits myself with a final inspection. The electrician quote was $1,150. Doing it myself looked like $400 in material plus a Saturday. On paper, easy. But my gut said I was missing something about the shelter's exposure to weather, and I overruled it because the cost difference was almost $700.
I pulled the permit, bought a 125-amp load center, mounted it, ran two circuits. Then it rained. Then the main tripped during a welding pass and kept tripping. Then the electrician I eventually called pointed out, gently, that I'd used an indoor-rated enclosure in a shelter with a missing north wall and no drip protection, that my grounding electrode conductor was undersized, and that the welder needed its own circuit rather than sharing with the compressor.
Per NFPA 70 (National Electrical Code), outdoor and damp-location equipment must be installed in listed enclosures rated for the environment. Wet-location ratings aren't optional. Verify the current edition requirements with your local AHJ — they vary by jurisdiction and edition adopted.
I'm not 100% certain the enclosure was the root cause of the nuisance tripping, but it definitely wasn't helping, and it wasn't legal. The rework invoice came to $620 on top of the material I couldn't return, and we lost the better part of a week.
Even after I approved the rework, I kept second-guessing. What if I'd just paid the $1,150 in September? The two weeks until the electrician could fit us in were stressful, mostly because I knew the answer.
The Tractor Supply run that wasted an afternoon and $210
Meanwhile the John Deere X350 mower was down on the other side of the shop — worn deck belt, blades that needed replacing, and a bagger kit we'd bought used that never did fit right.
I did what I've done a hundred times: I drove to Tractor Supply with the old belt in my hand, walked the aisle, matched length and cross-section, and bought what looked right. Same for blades. Both were "close enough."
Neither was. The aftermarket belt held for about four hours of mowing before it started slipping, and the blades had a different center hole than the spindle on that deck. Total damage: $210 across belt, blades, and a second trip, plus a day and a half of a mowing crew standing around.
What I should have done was order by the OEM part number off the deck serial plate. That's boring advice and it's the whole lesson. When you're looking at John Deere X350 attachments, the year and deck size matter more than the model name — Deere changed deck sizes across the X350 run, and the aftermarket catalogs don't always split them cleanly. The dealer counter looked it up in about ninety seconds and charged me maybe 15% more than the box store. Fifteen percent would have been cheap.
Where the forklift training went wrong
The shelter build meant moving pallets: the load center crate, the compressor, a pallet of block. We have a 5,000-lb warehouse forklift and one certified operator, and in late October he was out for two weeks.
I had a newer hire, Danny, who'd run a forklift at a lumber yard for a couple of years. I asked him to unload a truck. He was fine. He was also not trained on our truck, our yard, or our load charts, and I never put him through the program.
OSHA 29 CFR 1910.178(l) requires that powered industrial truck operators be trained and evaluated on the specific truck and conditions they'll operate in, and re-evaluated at least every three years. "He's done it before" is not a training program.
Third week of October, Danny took a pallet of block off the truck with the mast up higher than it should have been and no back tilt. The load tipped, took out a shelter door, and bent a hydraulic line hard enough to crack the fitting on the tilt cylinder. Nobody was hurt, which is the only part of this story I feel okay about.
Repair: $1,150 between the door, the hydraulic work, and a rental while ours was down. If I'd slowed down and written the training on how to operate a forklift into the schedule instead of treating it as overhead, that's a $1,150 check I don't write.
What it actually cost, and what we got
Roughly: $4,200 for the crawler, $690 in parts and rework around the shelter, $210 on the wrong mower parts, $1,150 on the forklift incident. The crawler itself needed maybe $340 in ignition and fuel work, which was the one line item that came in under estimate.
Call the avoidable portion about $2,300. The crawler finally graded its first lot in mid-December, not the following week.
I have mixed feelings about how this went. On one hand, we now have a working dozer for less than 20% of a new one, and the shelter is nicer than it was. On the other hand, I turned a two-week project into three months because I optimized three small purchases and ignored three bigger process questions. That's not a bad-luck story. That's a me story.
The checklist I built afterward
This is the part that actually changed. I stopped keeping the lessons in my head and put them in a shared sheet that the whole shop can see — one tab per machine, one row per part, with the OEM number, the serial number break, the source, the last price, and the date. It's not fancy. It's a spreadsheet with some dropdowns and conditional formatting. But in the past 18 months we've caught 47 potential ordering errors before they turned into returns, and that number is the only reason the owner didn't take purchasing away from me.
- Serial plate first. No order gets placed without a photo of the machine's serial plate in the thread. Model name is not enough, especially on mid-run John Deere equipment where deck sizes and part numbers changed.
- Anything electrical that touches an exterior wall gets quoted by a licensed electrician. I can still do the work if the quote is sane. I just can't decide from the aisle of a hardware store.
- Forklift certification is a schedule item, not a favor. No signed evaluation card, no seat time. That rule is now on a laminated card on the truck, mostly so I don't talk myself out of it at 4:45 on a Friday.
If you're running equipment purchases for a small operation, my advice is unremarkable: verify the part number, don't self-install electrical in a wet shelter to save $700, and treat operator training as a cost you budget for rather than a cost you find out about later. We spent $2,300 discovering that a spreadsheet would have been cheaper.