EKCO U122 - Restore or spares?

[Gabe001]

Lawrence I was reading some threads about this. Is there a limit to this capacitor size? The reservoir cap obviously has a size limit,but opinions seem to vary on whether bigger is always better with regards to smoothing caps, with the possibility of stressing the rectifier upon switch-on. Is this a theoretical or a real possibility?

[ms660]

Originally Posted by DonaldStott

To what extent is that related to the 'quality' of the local mains supply and house wiring?

Next on my list - just looking at the circuit diagram to decide where to fit the 220 Ohm and a 33uF? I'm assuming the 33uF is an electrolytic?

House wiring is just like another antenna, you only need 50Hz to modulate an RF carrier that's on the mains and there it is, the modulation might be done by another device on the same mains or by an oxidized connection in the mains system for instance.

The 33uF would be an electrolytic, remember not to connect the -ve to chassis but to the same line as the -ve of the reservoir capacitor which is the switched mains Neutral line, if the hum you are getting is due to or partly due to the HT ripple voltage then an RC filter should help to reduce it.

Basically because there's no ripple cancellation in the replacement transformer then a CRC filter (aka a pi filter) fed from the rectifier should help compensate for that, the first C is already there in the form of the reservoir capacitor hence just adding an R and C to complete the pi filter, you could use a choke instead of a resistor as that would perform better but finding a space for it might be a problem.

Schematic wise (Trader) cut out the short horizontal line that's connected to the +ve of C28 and replace it with a 220 Ohm resistor, then from the right hand side of that resistor connect the +ve of the 33uF capacitor and connect its -ve to the same line that the -ve of C28 is connected to.

Lawrence

[DonaldStott]

Thanks Lawrence, much appreciated.

I'll have a go at implementing the CRC filter (aka a Pi filter) over the weekend.

I've had a look at several online Pi filter calculators so would be interested in how you arrived at a 220 Ohm and a 33uF ?

[ms660]

Originally Posted by DonaldStott

I've had a look at several online Pi filter calculators so would be interested in how you arrived at a 220 Ohm and a 33uF ?

I just put them into a voltage divider calculator (R and Xc) that gave approx 3:1 reduction and approx 10 volts DC voltage drop at a load current of 45mA, a 470 Ohm resistor gave approx 5.8:1 reduction and approx 20 volts DC voltage drop, I didn't go for a lower voltage drop than 10 volts because I don't know what the max hot switching current limit of rectifier is, so on reflection maybe try with 470 Ohms instead of 220 Ohms and see what difference the lower HT makes to the performance, at the end of the day it's all a trade off when trying to reduce the ripple voltage with a CRC filter, less ripple = less DC volts available.

The initial 220 Ohm value was arrived at from a Hacker GP42 schematic.

Lawrence.

[DonaldStott]

I managed to lash-up a temporary Pi filter following Lawrence's detailed advice and guidance about where to connect the 220 Ohm resistor and 33uF electrolytic capacitor.

RESULT - the hum has all but disappeared and can only be heard very faintly when the volume control is turned to a minimum!

I'll review Lawrence's additional advice on options to fine tune the Pi filter "I didn't go for a lower voltage drop than 10 volts because I don't know what the max hot switching current limit of rectifier is, so on reflection maybe try with 470 Ohms instead of 220 Ohms and see what difference the lower HT makes to the performance ... " I'll carefully check the voltages on V5 and V4 for both options.

I am, however, very impressed and pleased with this outcome - something I've never used before!

So thanks again to Lawrence for this and for introducing me to the world of the Pi filter! 😁

[Gabe001]

Well done to you and to Lawrence for getting this sorted

[ms660]

What ever you do don't try and increase the value of the added capacitance or reduce the value of the added resistance, in fact the lowest value capacitor in conjunction with the highest value of the added resistor in terms of acceptable ripple hum and performance the better just in case the rectifier doesn't like it.

Lawrence.

[DonaldStott]

Checked the valve pin voltages with the 220 Ohm resistor and 33uF electrolytic capacitor- figures in brackets are from the Service Sheets:-

AC Mains Supply : 246V AC

V5 Rectifier UY41

Anode : 223V AC (215V AC)
Cathode : 244V (218V)

V4 Output Valve UL41

Anode : 212V (213V)
Screen : 75.8V (76V)
Control Grid : -ve 4.0V

Apart from a reduced bias on V4 (previously -ve 4.38V) these figures don't look too bad to me - I am of course using a DMM and not an AVO.

@Lawrence - you mentioned that I could " ... use a choke instead of a resistor as that would perform better" - any recommendations for a value and type?

[ms660]

Originally Posted by DonaldStott

@Lawrence - you mentioned that I could " ... use a choke instead of a resistor as that would perform better" - any recommendations for a value and type?

Something around 5 to 10H is often mentioned, the last one I used was somewhere around 10H as far as I can remember, it was the only one I had to hand, it was used in a PSU I built, the choke must be rated for the DC current at least, and suitable for the voltages encountered, ie: good insulation between the winding and core etc.

With a suitable choke you can use a lower value of filter capacitor compared to an RC filter to get the same ripple voltage reduction.

Lawrence.

[DonaldStott]

Didn't appreciate that these 10H chokes were so chunky, size and weight, so I think I'll just stick with the 220 Ohm resistor and 33uF electrolytic capacitor meantime. There is enough space under the chassis to fit a small tag strip to safely hold these components.

That about wraps up the electronics for this set which has been substantially rebuilt! Just pleased that I don't have to make an economic return on such projects! Thanks to everyone for their help and support, particularly Lawrence for his valuable contributions.

Now back to those cosmetic case and knob issues ...

[DonaldStott]

A quick follow up question as I'm very impressed with the impact of the Pi filter on the level of hum with this set.

Of course it got me thinking about the hum on my Bush DAC90A and to what extent a Pi filter with the same value of components would be as effective - 220 Ohm resistor and 33uF electrolytic capacitor. If these values aren't applicable then how would one go about calculating the correct values. Lots of online Pi filter calculators but not sure which ones I should trust?

[Gabe001]

The bush wasn't intended as a high end music centre. With this in mind, I do t think the hum in the bush is unacceptable.

I would leave it as is. If it's sound quality you're after, try a Murphy or a German set

[DonaldStott]

Originally Posted by Gabe001

The bush wasn't intended as a high end music centre. With this in mind, I do t think the hum in the bush is unacceptable.

Thanks Gabe - I've never expected hi-fi performance from these 1950's vintage valve sets but if the hum can be tamed, as I have done with the EKCO U122, then why not with other sets? It's all a voyage of discovery and part of my ongoing education!

[Gabe001]

Of course Donald. My comment appears a bit more confrontational than I intended. Apologies.

Did you manage to sort out bluetooth on the Bush?

[DonaldStott]

Originally Posted by Gabe001

Of course Donald. My comment appears a bit more confrontational than I intended. Apologies.

Did you manage to sort out bluetooth on the Bush?

No apology required Gabe - thanks for your ongoing support and interest.

I was hoping that Lawrence would perhaps give me a steer on this?

As for the Bush DAC90A Bluetooth Project, I'm hoping to draft an update in the next day or two.

[Gabe001]

Well, he did say

"the lowest value capacitor in conjunction with the highest value of the added resistor in terms of acceptable ripple hum and performance the better just"

Depending on how healthy your rectifier is, if you really want to go down this route (it's your radio), why don't you try a 330 resistor and 10uf cap as an experiment?

[ms660]

Originally Posted by DonaldStott

I was hoping that Lawrence would perhaps give me a steer on this?

I'm not into circuit modification as such unless it's for hard to get bits or bits unobtanium.

That said, rectifiers have limits, the UY41's limits haven't been reached in the Bush DAC90a circuit, see if you can spot which limits haven't been reached, one in particular can have an effect on the ripple voltage.

Lawrence.

[DonaldStott]

Originally Posted by ms660

...rectifiers have limits, the UY41's limits haven't been reached in the Bush DAC90a circuit, see if you can spot which limits haven't been reached, one in particular can have an effect on the ripple voltage.

Found this in an earlier Thread:-

"I generally reduce the hum level on Bush DAC90A receivers by shunting the reservoir capacitor with a further 32 uF. This is well within the limit for the UY41 at the current drawn by this receiver."

[ms660]

That's the one, the reservoir limit that is.

Lawrence.

[DonaldStott]

The Valve Museum page states:-

'To limit the current to a safe value for the UY41 the series resistance needs to be at least 210 Ohms. The maximum reservoir capacitor must be limited to 50 μF'

Looking back at the quote in my earlier Post how would one shunt the reservoir capacitor with a further 32 uF and not exceed the maximum? Not sure of the current drawn by the Bush DAC90A?