Version 1.32
21
Additional Notes for High mains supplies (From our customers)
HT supplies
The aim was to maximise the HT to the main output valves with the lowest series
resistance before X5, without introducing 100Hz hum, or resulting in a voltage on the
first reservoir capacitor (C7/8) greater than about 490Vdc (given their rating of
500Vdc). If the supply is a very healthy 240Vac mains supply and the calculated
value off load on the first set of capacitors was over 520Vdc; first power up measured
the voltage at X6 of around 492V, and it would have been higher at C7/8 but I didn’t
measure it).
I want to be able to swap valves about, so the design has to cater for the different
cathode bias currents for each set/type; I can’t optimise for a given set, only
compromise on values of series resistance that don’t generate too low a voltage on
sets with higher cathode current or too high a voltage for sets with a lower cathode
current.
Major factors impacting this are the cathode currents in V1/V2 and the actual voltage
of the incoming 240Vacmains.
Swapping sets of V1/V2s;
using a matched quad of EH-EL34s or a matched set of EH-6CA7s sees different bias
currents varying from 69.6 to 77.9mA. These different cathode currents result in
different voltage drops across the 50R resistor I have added, R35 and R52 and so
different HT supply values at the anodes (which in turn lead to alterations in the bias
current until it all balances out!)
I guess I could go as far as to add a variable resistor in series with each 470R resistor
(100R variable in series with 3k9 2W across the 470R; R = 419R to 519R) to balance
them all out and set them the same for each valves in each set I plug in..but a lot of
effort to go to, given that I love the sound I’m getting now!
Mains variation:
Which I can see indicatively by measuring the heater ac volts to V3/V4 (more on
heaters later!) this varies between 6.44 and 7.05Vac.. I haven’t measured the
incoming mains voltage at the same time to verify if it is fluctuations in the 240Vac
supply or if it is down to load regulation on the torroidals; I suspect predominantly
variation in the input depending on the time of day.
Modifications to the circuit, see last page:
The 50R resistor between the bridge and the first cap is very effective since it knocks
the peak currents down and reduces the 100Hz peak ‘ac’ voltage in to C7/8.
Changing R52 from 100R to 50R between X6 and X5 also helped..
With these two changes I have a V1/V2 anode supply at X5 between 447 and
465Vdc..depending on the valves fitted 6CA7s or EL34s. There is no discernible hum
on the output (practically none can be heard at the speakers).
Heater voltages
Due to a healthy dose of 240Vac at the inputs; the heater voltages on the valves were
all a little on the high side 6.9 to 7.05V measured..This mod would be recommended
for anyone in the UK or anyone who has a healthy 240Vac mains supply.
While the 6N1P has a wide range of heater voltages from 5.7 to 7.5V the E88CCs and
ECC88s state 6.3 +/- 0.3, with a Vh max of 6.9V – as do the EL34s and the 6CA7s a
max of 7.0V.
Small resistances have been added in series with the heater ac supplies:
