Skip to content
Honest MileFix what is broken. Explain what is not.

Four repairs in full, including an expensive one

A workshop page that only ever ends in a small bill is not a record, it is a different sales pitch. The third case on this page is a ฿13,257 one.

HM-2411-038Honda City 1.5 · 2016138,200 km

Another shop quoted about ฿24,600 for the whole front end. We replaced one ฿890 link.

The owner said

A knock from the front over speed bumps, worse first thing in the morning. The owner said another shop had quoted for the whole front suspension, somewhere over twenty thousand baht.

What we did to find out

  1. 1Chai drove the test loop with the owner in the car — 0.4 km, three speed bumps — so that we were certain we were hearing the same noise.
  2. 2On the lift, a pry bar on every front chassis joint, each one compared left against right.
  3. 3Both shock bodies checked for an oil film.

What we found

  • Repair nowLeft front stabiliser link4 mm of play at the ball socket, audible when rocked by hand.
  • Plan itFront lower arm bushes, both sidesSurface crazing, not split. Recheck in 10,000 km.

What the owner approved

  • Labour, 1.0 h฿400
  • 555 / Sankei stabiliser link, left฿890
  • Materials฿80
Total, VAT included฿1,465.90

Offered and declined

Replace the right-hand link at the same time฿1,090

We offered it because both sides are the same age, but wrote on the estimate that the right side has no play today. The owner chose one side.

A gloved hand grips the old stabiliser link under a raised car; its ball-joint boot is split open and greasy
Before
The same frame and the same light, with a new black stabiliser link fitted, a fresh nut, and the area wiped clean
After

The result

Road tested on the same loop, over the same three speed bumps. The noise was gone. Six months later, at the next service, the right link still had no play.

The five things we checked and did not replace are on the same estimate, at ฿0.

Checked and not replaced

฿0
  • MonitorRight stabiliser linkNo play.
  • Plan itFront lower arm bushesCrazed, not split. Recheck in 10,000 km.
  • MonitorFront strutsDry, no oil film.
  • MonitorLower ball jointsNo measurable play.
  • MonitorTie rod endsNo play.

HM-2503-011Toyota Vios 1.5 · 2014156,700 km

The code named a ฿4,850 sensor. The cause was a ฿280 hose.

The owner said

Check engine light on, a slightly rough idle, and fuel consumption up by about 1.5 km per litre.

What we did to find out

  1. 1Scan returned P0171 — system too lean, bank 1.
  2. 2Live data: short-term fuel trim at +22% at idle, but only +4% at 2,500 rpm.
  3. 3That pattern is an unmetered air leak, not a failing sensor: a bad sensor is wrong by the same amount at every engine speed.
  4. 4Smoke test of the intake: smoke came out of a split PCV hose at the elbow.

What we found

  • Repair nowPCV hoseSplit at the elbow, about 12 mm, visible under smoke.

What the owner approved

  • PCV hose฿280
  • Labour, 0.75 h฿300
  • Materials฿40
Total, VAT included฿663.40
Thin smoke drifting out of a split in a PCV breather hose in a dusty engine bay
Before
The same angle, with a new matt-black PCV hose and a fresh clip, and no smoke
After

The result

After the repair, short-term trim came down to +3% at idle. The code was cleared and the owner drove 200 km; the light stayed off. The ฿600 diagnostic fee was credited in full.

A code tells you what the computer noticed. It does not tell you which part is broken.

Checked and not replaced

฿0
  • MonitorUpstream oxygen sensor — the part the code names. ฿4,850 fittedSwitching normally once the leak was fixed.
  • MonitorMAF sensorReading correct at idle and at 2,500 rpm.
  • MonitorThrottle bodyCleaned during the smoke test anyway. Not charged.

HM-2506-024Mitsubishi Pajero Sport 2.5 · 2013187,400 km

Sometimes it really is the whole job — ฿13,257 of cooling system, with a reason on every line.

The owner said

Needs topping up every three weeks. No puddle visible underneath. The temperature gauge is still normal.

What we did to find out

  1. 1Pressure test at 1.1 bar: it held for eight minutes, then started to drop.
  2. 2UV dye traced it to the water pump weep hole — which means the pump’s bearing seal is going.
  3. 3The thermostat housing gasket was weeping, the upper hose was soft and swollen, and the radiator side-tank seam had a crust on it.

What we found

  • Repair nowWater pumpLeaking from the weep hole — that is the bearing seal, not a gasket.
  • Repair nowRadiatorCrust at the side-tank seam, and pressure still dropped after the pump was temporarily blanked.
  • Plan itThermostat and housing gasketWeeping — and underneath parts that have to come off anyway.
  • Plan itUpper and lower hosesUpper soft and swollen; lower still firm but the same age.

What the owner approved

  • Labour, 7.5 h — drain, water pump and gaskets, thermostat, radiator, both hoses, vacuum fill, bleed, 40-minute road test฿3,000
  • Water pump, Aisin฿2,450
  • Thermostat, Aisin฿780
  • Radiator, Koyorad฿3,900
  • Upper hose฿420
  • Lower hose฿460
  • Gasket set฿180
  • Coolant, 6 L฿1,080
  • Materials฿120
Total, VAT included฿13,257.30
A UV torch picking out a coolant trail running down from the water pump under a raised pickup
Before
The same angle, with a clean new aluminium water pump, a new hose, and no trail left
After

The result

Re-pressure tested at 1.1 bar and held for the full twenty minutes. Temperature stayed flat through a 40-minute road test that included twelve minutes in traffic.

We said do it now because the pump was leaking from the weep hole — that is a bearing — and the same belt drives it. If the bearing seizes the belt goes, and on this engine that becomes a valve-and-piston repair. The estimate we did not want to write is the one after that.

Checked and not replaced

฿0
  • MonitorRadiator capTested and held 1.1 bar.
  • MonitorFan motor6.8 A draw, in spec.
  • MonitorCoolant reservoirNo cracking.
  • MonitorHeater coreFlow and inlet-to-outlet temperature difference both normal.
  • MonitorDrive belt41,000 km on it, no cracking. It has to come off for the pump anyway, so refitting it costs nothing. A new one would have been ฿890 and we did not think it was needed.

HM-2505-007Isuzu D-Max 2.5 · 2014209,800 km

Two refrigerant top-ups in a year is not a repair.

The owner said

The air is cool but not cold, and worse in traffic. Another shop added refrigerant twice last year.

What we did to find out

  1. 1Vent temperature 18.5 °C at idle with 34 °C ambient. It should be 8 to 12 °C.
  2. 2Low side 15 psi, high side 165 psi — both low, which means the charge is down.
  3. 3UV dye, twenty minutes running, then a blacklight: a seep at the condenser-to-line O-ring, plus older dye residue underneath the condenser from the previous top-ups.

What we found

  • Repair nowCondenser-to-line O-ringSeeping, clearly under UV. This is where both previous charges went.
  • Repair nowReceiver/drierNot because it failed — because the system is being opened, it has absorbed moisture, and it cannot be reused.

What the owner approved

  • Labour, 2.75 h — recover, replace O-rings and drier, evacuate 45 min, recharge by weight, retest฿1,100
  • O-ring and seal kit฿260
  • Receiver/drier฿780
  • R134a, 550 g฿660
  • PAG 46 oil฿180
  • UV leak-trace dye฿150
Total, VAT included฿3,349.10
Yellow-green leak-detection dye glowing around the air-conditioning line joint at the condenser, under UV light
Before
The same joint in ordinary light, with a new O-ring and bolt, wiped clean with no dye left
After

The result

Vent at 7.8 °C at idle and 6.9 °C moving, with 33 °C ambient. Pressures 28 and 210 psi. The ฿700 diagnostic fee was credited in full.

Refrigerant is not a consumable. If it is low, it went somewhere. The job is to find where — not to fill it up again.

Checked and not replaced

฿0
  • MonitorCompressorClutch gap 0.45 mm, in spec. No bearing noise.
  • MonitorCondenserFins undamaged, no leak in the core itself.
  • MonitorCondenser fan6.2 A draw, in spec.
  • MonitorExpansion valveSuperheat normal after the recharge.
Book an inspection