Recently I acquired a second-hand and untested TF2700 LCR bridge from Marconi. This is a fantastic piece of engineering and a superb LCR bridge that can only be beaten by an expensive impedance analyser, something I can clearly not afford.
This bridge is perfect for selecting accurately capacitors for a RIAA network with extreme accuracy. In fact, the 1% precision can be improved down to 0.25% by replacing the main bridge 0.5% resistors with 0.1% ones as suggested by Morgan Jones in his “Building Valve Amplifiers” (Second Edition) page 330 and 331.
I was looking to build my own LCR bridge but given the Marconi’s circuit is so simple and has a great reputation, I decided to venture on the repair of this old unit.
Firstly, you have to replace the PP9 battery connector with either a six AA battery pack or a PP3. I added a PP3 connector for a rechargeable battery. You have to clean all the switches and parts. This takes a significant part of time, but is worth doing it.
The old electrolytic capacitors must be replaced. Is not a great deal but you can do that very easily without dismounting the PCB boards:
Newer capacitors are much smaller in size. Despite being axial types, the voltage requirements are very low (16V) so there are still easy to find:
Once all electrolytics have been replaced, you can proceed to test in isolation the 1kHz oscillator. You can do this by measuring with the scope the output of the T1 secondary. You wouldn’t expect any calibration needed unless C1 or T1 are damaged.
You don’t have to upgrade the range multiplier 0.5% resistors. But doing so, will improve the accuracy of the bridge down to 0.25%. These days, 0.1% metal film resistors are not that expensive. You have to get a set of 1M, 100K, 10K, 1K and 100Ω ones but the 10Ω one should be a non-inductive 0.1% one:
Hopefully the standard 100nF 0.1% capacitor is ok, otherwise getting hold of one of those it may prove to be real headache. The real challenge are the potentiometers. I was lucky that the balance (fine) RV1 wirewound potentiometer was in good condition, but unfortunately discovered that the sensitivity pot RV2 (dual gang) was seriously damaged. I was testing the detector amplifier when I found that the error signal wasn’t amplified by the second stage transistor (VT3). So dismantled this potentiometer and tried to repair it unsuccessful as part of the sliding arm carbon broke when I was cleaning it:
The end result was that I had to put on hold this project until I could source a 2K5/25K LIN/LOG ganged potentiometer. I could have attempted in building this from a ganged 2K5 and 25K pair but I was keen to get this pot somehow.
Luckily someone on the forums recommeded trying Blore Edwards potentiometers from Wales.
I was surprised to see that I could specify the potentiometers for a custom build at a very reasonable cost. Within a week I had a pair of these ganged potentiometers with me ready to replace RV2.
I completed the refurbishment successfully and re-calibrated the unit by using a ladder of 100K/200K 0.1% resistors as recommended by Morgan Jones in his book.
The unit now works like a charm, it’s stable from initial power up and can accurately measure capacitors and even chokes/primary inductance of SE transformers using an external CCS and coupling capacitors. Easily to set up and drive with an external oscillator to ensure the inductance is measured at correct operating voltage.
I’m planning to refurbish a second TF2700 I have in the future if you are interested please let me know.
cheers
Ale
The failed fine balance pot in my TF2700 is a 110 Ohm wire-wound one made by Clarostat. Another 2700 that I own has a Colvern 110 Ohms component.
It looks as though the cause of the pot’s going open circuit was that it had suffered from damp at some time before the bridge came into my possession as there were signs of green corrosion (verdigris?) on the track. I tried to solder the breaks but was worried that the former on which the track is wound would break up. The breaks were on the edge of the track away from the wiper, which makes me even more suspicious that damp was the culprit as that edge lies in the corner of the pot body away from any ventilation that would otherwise have caused it to evaporate.
I remembered from years ago that one could get a silver loaded ‘paint’ for repairing the car rear window demister elements that had been damaged – the surface type element not the ones embedded in the glass – and was lucky enough to buy a small syringe of it off Ebay for £1•16 included postage, it was sold as a PCB repair material.
I put the smallest blob that I could on each break and checked the track’s continuity , sadly it was still open-circuit so I packed up for the evening, intending to try again the following day. To my relief when I checked the next morning the track read 112 Ohms end-to-end, obviously with hind-sight I’d needed to let the solvent evaporate. I reassembled the pot using 3 millimetre screws, washers and nuts in place of the rivets that originally held the track in place and checked that the wiper made contact throughout its travel with no sign of intermittent contact.
On fitting the pot back into the bridge I was gratified to find that, having replaced the ancient electrolytic capacitors (of course!), everything worked as Signor Marconi originally intended.
It is a tricky job getting the track out with wrecking it but with care and patience it can be done transforming what had become simply a door-stop into an accurate and versatile bit of test equipment and giving one a smug feeling of satisfaction!
Good luck if you decide to try the repair though if the pot has failed due to wear by the action of the wiper on the track I fear that my solution simply is not applicable.
I’ve now (end July 2014) put it on Ebay.