wd hd3 problem

[Princo]

sorry, meant mV, not mA

[Richard]

Originally Posted by Princo

The measured voltages with the cap C6 disconnected from pin 9/R8 are:

R Chan Pins L Chan

1 110v / 105v
2 0v / 0v
3 0.9v / 0.9v
4 H 6.3ac/ H
5 H / H
6 195v / 196v
7 14.8v / 15v
8 196v / 196v
9 0v / 0v

There is very small voltage passing C6 fluctuating between +100mA to -50mA, this is a new cap.

All as they should be.

Excellent figures for all stages with the cap disconnected but C6 seems to be passing something. Is your mA figure a typo, was it mV? C6 shouldn't be passing any DC, mA or mV. Check with the meter on mV setting with the black probe connected to signal ground (ie neg of C3), the red probe on the free end of the C6, and the other end of C6 still connected to Pin1 100V dc. What cap is it, plastic or paper, what voltage and capacity value?

[Richard]

A cap may pass a fluctuating signal if the DC is not smooth. The Pin1 supply is common to both channels but only one is suffering, so that points to a poor connection between C6 and Pin1 or Pin1 and the PSU/C3 positive.


Another thought, check both Pin1s are connected to the positive of the last of the smoothing caps (C3), ie after 4k7 and not before as DC would be rougher there.

[Richard]

As the reading on Pin 9 is fluctuating try with the meter set for AC volts. It could be it’s trying to read as DC and can’t lock on.

[Princo]

Ive tried both auricaps 0.22mf 600v pp and 0.22mf orange drops (new), both give roughly the same voltages when connected or pin 9 leg disconnected, as i posted earlier.

When disconnected all voltages are spot on on all pins, there is a small fluctuating dc voltage of +/- 30mV at the disconnected leg of C6, if i switch to ac reading is steady 40mVac, however did the same on the Rchan for comparison and it is the same.

Have checked again all connections around pin 1/C6/R4/C3.

I disconnected R11 completely (load resistor Rchan), same result.

Here is the last round of voltage checks with all connected and dummy load resistors fitted.

ON-LOAD

RChan LChan

1 108v / 173v
3 0.9v / 0.2v
6 191v / 186v
7 14v / 22v
8 191v / 184v
9 0v / +/- 4mV

This is a change since earlier, i connected dummy load to both channels.


So now on-load anode voltage pin 1 much to high on triode stage.
Bias voltage on pin 3 too low

Bias voltage on pin 7 has gone from 2v no-load to 22v on-load.

NO LOAD

R Chan /L Chan
1 110v /107v
3 0.85v/ 0.9v
6 196v /205v
7 14v /2.0v
8 197v/ 205v
9 0v/ -47v


Running out of things to try

[Richard]

The erratic readings are strange. Do you have another pair of valves to try? Are the valve pins clean? What construction is the build, are the valve bases on a pcb or chassis mounted and are they making good contact with the valve pins?

The figures in red suggest the triode stage is failing to conduct; voltage is not being dropped across plate resistor R4/11 nor across its cathode resistor R9/16 so replace them.

Chase the measurements back and meter at the positive of C3. Replace the wires between C3 and Pin 1. Replace R4/11 100k. The R channel is fine so just keep checking and changing parts of the L channel.

[Princo]

Hi Richard

Thanks for taking the time to help i will try as you suggest, its very strange as the R chan is a carbon copy of the L chan, i thought maybe I had transcribed the schematic incorrectly or made an earth somewhere i shouldn't.

Ive ordered a PCB from Matthew and will use that, but i will try and sort this hard wire version first.

I'll let you know if i find anything

Cheers
Jim

[bob orbell]

Jim wrote - but i will try and sort this hard wire version first. - I did mention this in post 15, I think Jim should have mentioned it earlyer. Bob

[Richard]

The reason for asking is that hardwiring and pcb methods both have pro’s and cons.

The HD3 pcb has the PSU onboard with connections made by tracks so, as long as components are correct and orientated correctly, it should be easy. However, a break in the tracks is sometimes difficult to spot, a fault with a valve base is harder to see or clear, and dry joints from thick component leads being soldered into thin tracks with insufficient heat not allowing the joint to wet and flow is common (apply the iron and solder to the thick part first then slide it along onto the thin track and watch it flow).

Hard wiring is nice but there is more room for error in making the correct connection or signal/noise being induced from nearby wiring with loose wires and poor layout.

With valves bases, fit them to a valve before fitting to make sure all connections are good. Larger octal ones can be quite tight and need a little easing to get them to fit nicely. With pcb bases I fit a valve in the base whilst soldering to make sure the pins and base connections are aligned as there is quite a lot of movement. If the job has already been done then you can still fit a valve then remake each solder joint to allow the connector to move a little if it needs to afterwards. Chassis mount valve bases are better in this regard as the connectors should still float in the ceramic after being soldered to the wires.

Beware new old stock valves which may need the pins cleaning with a glassfibre pencil or similar. A little blue oxidation is not usually a problem but heavy black crud on small signal pins can cause noise.

[Princo]

Hi Richard

Thanks for the advice, good points.

I will complete the pcb version when i get the board and do a hard wire version later.

I did a few more checks last night with feedback connected /disconnected.

With fb:
pin 1 110v
pin 3 1v
pin 7 0.4v
pin 9 -51v

So triode stage working, pentode stage fault

without fb:

pin 1 203v
pin 2 7v
pin 3 14v
pin 9 0v

So now triode stage not working and pentode stage working

Anyway i pulled the whole thing apart again but i would just like to know what is going on.