LT regulator in the TF144H Sig Gen - UK Vintage Radio Repair and Restoration Discussion Forum

Silicon · 19th Oct 2018, 10:00 pm

I understand the operation of the HT regulator in the Marconi TF144H/4 Signal Generator.

A potential divider takes a sample of the output from the HT regulator and compares it to a voltage reference. The comparator then sends a control signal to the control element (the EL81 valve) to correct any variation from the required output voltage.

At first I thought that the 6.5V LT regulator worked the same way except that it takes a sample of the HT voltage and not the LT output voltage. This does not make any sense to me.

Could it be that the comparator transistor VT203 compare the output voltage which is at its emitter, with a reference voltage at its base?

This would imply that the potential divider R214, RV202 and R215, which 'samples' the 250V HT, is acting as a voltage reference.

A failure of the HT regulator will cause a failure of the LT regulator.

Could this circuit be improved by adding a voltage doubler to the LT1 winding to create a +26V supply. This could then power a 9V zener which would be the new reference voltage for the potential divider.

turretslug · 19th Oct 2018, 10:21 pm

It always struck me as odd, too, or at least skating on unnecessarily thin ice. I'm prepared to accept that there were some very clever folk at Marconi who didn't always think along conventional lies and perhaps there was some subtlety about allowing HT and LT to drift in track- though the catastrophic event of EL81 being driven into saturation by circuit failure then becomes doubly catastrophic with both metering thermistor and valve heaters in peril, with boosted emission from a 5673 running with 400V-odd HT! Maybe LV Zeners were expensive and exotic at time of design and the budget was stretching.

I'm sure that something like a TL431 or 5.1 to 6.2V Zener (the most temperature stable region) could be accommodated with a few suitable changes to circuit constants.

19th Oct 2018, 10:00 pm	#1
Silicon Nonode Join Date: Jun 2007 Location: Coulsdon, London, UK. Posts: 2,163	LT regulator in the TF144H Sig Gen I understand the operation of the HT regulator in the Marconi TF144H/4 Signal Generator. A potential divider takes a sample of the output from the HT regulator and compares it to a voltage reference. The comparator then sends a control signal to the control element (the EL81 valve) to correct any variation from the required output voltage. At first I thought that the 6.5V LT regulator worked the same way except that it takes a sample of the HT voltage and not the LT output voltage. This does not make any sense to me. Could it be that the comparator transistor VT203 compare the output voltage which is at its emitter, with a reference voltage at its base? This would imply that the potential divider R214, RV202 and R215, which 'samples' the 250V HT, is acting as a voltage reference. A failure of the HT regulator will cause a failure of the LT regulator. Could this circuit be improved by adding a voltage doubler to the LT1 winding to create a +26V supply. This could then power a 9V zener which would be the new reference voltage for the potential divider.

19th Oct 2018, 10:21 pm	#2
turretslug Dekatron Join Date: Nov 2011 Location: Surrey, UK. Posts: 4,396	Re: LT regulator in the TF144H Sig Gen It always struck me as odd, too, or at least skating on unnecessarily thin ice. I'm prepared to accept that there were some very clever folk at Marconi who didn't always think along conventional lies and perhaps there was some subtlety about allowing HT and LT to drift in track- though the catastrophic event of EL81 being driven into saturation by circuit failure then becomes doubly catastrophic with both metering thermistor and valve heaters in peril, with boosted emission from a 5673 running with 400V-odd HT! Maybe LV Zeners were expensive and exotic at time of design and the budget was stretching. I'm sure that something like a TL431 or 5.1 to 6.2V Zener (the most temperature stable region) could be accommodated with a few suitable changes to circuit constants.