EMI RS124

[Rob Flinn]

That looks correct BUT with the input pot fully clockwise or turned all the way up,  If the pot was fully anti clockwise both wipers would be towards the centre tap. 


The thin sound you are describing still sounds to me like the output is one legged. i.e you only have a hot or cold hooked up on the i/o connections, or possibly only one half of the balanced circuit is working.    Check that you have all your transformer connections hooked up properly preferably with the continuity setting on your meter.  If they all check ok, then take some voltage readings on the valves.  Check that both anodes on the same valve have very close to the same voltage to ground.

i.e on V1 pins the anodes 1 & 6 should be around 75v if you are using a 6bc8 or 55v if using an ecc189.  The anodes on V2 (pins 1 & 6) should be around 145v.

 

[letterbeacon]

That looks correct BUT with the input pot fully clockwise or turned all the way up,  If the pot was fully anti clockwise both wipers would be towards the centre tap. 

The thin sound you are describing still sounds to me like the output is one legged. i.e you only have a hot or cold hooked up on the i/o connections, or possibly only one half of the balanced circuit is working.    Check that you have all your transformer connections hooked up properly preferably with the continuity setting on your meter.  If they all check ok, then take some voltage readings on the valves.  Check that both anodes on the same valve have very close to the same voltage to ground.

i.e on V1 pins the anodes 1 & 6 should be around 75v if you are using a 6bc8 or 55v if using an ecc189.  The anodes on V2 (pins 1 & 6) should be around 145v.

Great thanks, I just wanted to rule out the input pot being wired wrong.

All my voltages on my previous point to point version were spot on, but it appears they're slightly different now since I've rebuilt it on a tag board. A bit higher than before.

I'm planning to take a closer look tonight and will report back. Also planning to get the oscilloscope out and tracing the signal through the amp.

 

[Rob Flinn]

I just got the mains transformers through today for my 2 units. I used a pair of R core transformers that were the cheapest thing I could find to do the job, 57Euros for the pair includung shipping fro France.  They have a 240v secondary, but I found a 1k 7w resistor bring the h.t down almost perfectly & doesn;t seem to get too hot,  They are now both running without the bench PSU.  I was able to run some stereo program through them tonight. I really  like the sound of these boxes, they sound sweet .  I'm debating whether to add an attack control.  I might experiment with it tomorrow. 

 

[letterbeacon]

I just got the mains transformers through today for my 2 units. I used a pair of R core transformers that were the cheapest thing I could find to do the job, 57Euros for the pair includung shipping fro France.  They have a 240v secondary, but I found a 1k 7w resistor bring the h.t down almost perfectly & doesn;t seem to get too hot,  They are now both running without the bench PSU.  I was able to run some stereo program through them tonight. I really  like the sound of these boxes, they sound sweet .  I'm debating whether to add an attack control.  I might experiment with it tomorrow.

Wow, that's a great price for 2 -nice work!

I've spent some time with mine tonight and I'm completely stumped!

I found a dodgy wire that meant the side chain wasn't getting to V1. I've fixed that and now it compresses, although it clicks a lot on the attack. I've got V1 balanced perfectly with a DMM so I don't know what it's doing that. But to be honest I'm sort of parking that for now because I want to focus on getting the amp working. So I've pulled the 6AL5 again for now.

I found that positive was shorting to the shield on the output XLR, so I've fixed that and now I'm not getting that weird garbled noise any more - just the usual very thin tinny sound. So basically back to how it was when it was a point to point build!

I ran some 1kHz tone through it and used an oscilloscope to trace the signal through the amp, and it's all there from the XLR input socket to the output socket. Of course, this doesn't tell me the point when the high pass filter comes in, but at least I can see that both legs of the push pull circuit seem to be passing signal.

Thinking out loud here, if my new tag board layout has the same problems as the point to point version, then that points to a problem with something not on the tag board. I've checked all the tube pins, made sure there are no shorts between the + and - on the audio cables...

One other thing to note is that before all my voltages were spot on with the schematic, but since building the amp on the tag board, they've all gone up by about 10v. Still within spec for an old tube amp (especially as I'm using a mixture of metal film and carbon comp resistors), but slightly odd nonetheless!

 

[Rob Flinn]

I would say that if you are seing a good signal on both halves of each valve & you are still getting a tinny sound it still is potentially that you have the output one legged.  Have you put a meter on resistance across pins 2 & 3 of the output xlr ?

You could also see where the F response goes by apllying a LF since wave & looking at that on your scope to see if the level is different to when you have 1K tone. 

 

[letterbeacon]

I would say that if you are seing a good signal on both halves of each valve & you are still getting a tinny sound it still is potentially that you have the output one legged.  Have you put a meter on resistance across pins 2 & 3 of the output xlr ?

With the output attenuator fully counter-clockwise I am measuring 389r between pins 2 & 3 on the output XLR. I've gone back to the 1k pot + 2 300r resistors as the output attenuator, by the way.

You could also see where the F response goes by apllying a LF since wave & looking at that on your scope to see if the level is different to when you have 1K tone.

That's a good idea, I'll have to try tomorrow now though.

 

[Rob Flinn]

With the output attenuator fully counter-clockwise I am measuring 389r between pins 2 & 3 on the output XLR.

If you have the 1k pot attenuator you should have basically a dead short across 2& 3 with the pot turned all the way down.  Just to check, you have the o/p transformer feeding the 2 x 300R & the pot is strapped across the non transformer side of the 2 x300R which also feeds the XLR pins 2 & 3
?

 

[letterbeacon]

If you have the 1k pot attenuator you should have basically a dead short across 2& 3 with the pot turned all the way down.  Just to check, you have the o/p transformer feeding the 2 x 300R & the pot is strapped across the non transformer side of the 2 x300R which also feeds the XLR pins 2 & 3
?

Sorry, I meant I'm measuring 389r between pins 2 & 3 when the pot is fully clockwise.

Yes I have it wired up as you describe.

 

[letterbeacon]

So I've had a chance to run some tone at different frequencies through the unit, and I think the problem is the transformer.

Running a 1kHz sine wave at -10dBFS into it and I measure about 2Vpp on the hot and the cold of the XLR in

The signal goes into the input transformer and I measure .2Vpp on the output of the input transformer.

Running a 100Hz sine wave at -10dBFS into it and I measure about 2vpp on the hot and the cold of the XLR in.

The signal goes into the input transformer and I measure .05Vpp on the output of the input transformer.

Unless I'm reading that wrong, it looks like the hi pass filter effect is coming from the input transformer, which is a bit of a bummer.

I don't know much about converting into dB, but according to this website  that's means it's 10dB quieter at 100Hz than it is at 1kHz. That looks like a high pass effect to me!

It's a Peerless 4651A, so it's old, so maybe it's defective. Unless it could be anything else? I've rechecked the wiring round there and it seems correct.

 

[Rob Flinn]

It's a Peerless 4651A, so it's old, so maybe it's defective. Unless it could be anything else? I've rechecked the wiring round there and it seems correct.

It seems a bit weird that the correct transformer for the unit has such a poor repsonse.

One thing that is odd is that the transformer is a step up transformer & you seem to be losing level.  Are you absolutely sure you have it wired up correctly ?  I don't have the spec for the transformer but does need connections between, for example 2 pairs of primaries.  e.g If it has 2 primaries do they need a link to put them in series ?

If it is definitely wired up correctly I would  check the transformer for continuity ?

 

[letterbeacon]

I have wired it up according to the 435 schematic on the Sowter website: http://www.sowter.co.uk/schematics/alt436c.pdf

So blue to pin 2 and red to pin 3 on the primary side, and then green to pin 1 of one pot on the dual pot, yellow to pin 1 of the other pot and the black CT going to both of the pot's pin 3s.

There's no continuity between the primaries and no continuity between the secondaries or the centre tap, which is good, I think? I'm not very good with transformers I'm afraid.

Out of interest, what transformers did you use?

 

[Rob Flinn]

and then green to pin 1 of one pot on the dual pot, yellow to pin 1 of the other pot and the black CT going to both of the pot's pin 3s.

If you take pin 1 of the pot to be the one the wiper goes to when you turn the pot down, The you should have the centre tap connected to both pin 1's. 

Pin 3 is what the wiper goes to when the pot is fully turned up. The green & yellow should go 1 each to the pin 3's.


You should be able to measure some resistance between blue & yellow. 

You should also be able to measure resistance between green & yellow.    If you measure resistance between the secondary centre tap & green, & also measure a fairly similar resistance between the centre tap &  yellow.

Out of interest, what transformers did you use?

I used a pair Gardners transfomers I had lying around for the input & Sowters for the output.

 

[letterbeacon]

Sorry yes I have the pot connected as you described - CT to pin 1, green and yellow to pin 3.

I measure 2.13k between green and yellow, but no resistance between blue and yellow.

I measure 1.13k between yellow and black CT and 998r between green and black CT.

Looks like there is no continuity or resistance between primaries and secondary...

Is it worth me testing the frequency response of the input transformer? Presumably it's not as simple as hooking REW up to it, as I'd need to simulate the impedance of the circuit after it, right?

 

[Rob Flinn]

You shouldn't have any continuity between primary and secondary they should not be connected because the are 2 separate insulated coils of wire,.

Do you show any resistance between the two primary connections, blue and red ?

 

[letterbeacon]

You shouldn't have any continuity between primary and secondary they should not be connected because the are 2 separate insulated coils of wire,.

Do you show any resistance between the two primary connections, blue and red ?

Yes, it shows 609r between red and blue.

 

[snod_donkey]

Hi, i am looking at having a go at building a couple of these up. Are you chaps using the circuits as published on the Sowter page?  I have seen a few floating about but was not sure which had been known good workers

Any pointers would be much appreciated

 

[Rob Flinn]

Yes, it shows 609r between red and blue.

That possibly about right.  From the information you are giving I would think that the transformer is ok.

You could probably just hook it up straight to REW for a basic test to see what your getting.

Can you post a close up photo of your input pot, so I can have a look at theway it's wired up ?

 

[letterbeacon]

That possibly about right.  From the information you are giving I would think that the transformer is ok.

You could probably just hook it up straight to REW for a basic test to see what your getting.

Can you post a close up photo of your input pot, so I can have a look at theway it's wired up ?

Before I read your post I decided to take the input pot out of the equation and hooked up a couple of resistors as described in the diagram attached and I'm still getting the super-tinny audio out.

I'll hook up REW to the transformer this evening and will see what I get. I worry though that if I haven't got the impedance exactly right, it might not be a true frequency response, or maybe I don't have to worry about that?

 

[Rob Flinn]

Before I read your post I decided to take the input pot out of the equation and hooked up a couple of resistors as described in the diagram attached and I'm still getting the super-tinny audio out.

I'll hook up REW to the transformer this evening and will see what I get. I worry though that if I haven't got the impedance exactly right, it might not be a true frequency response, or maybe I don't have to worry about that?

It probably won't be that far out because the input impedance of your audio interface will  be 10's of k

 

[letterbeacon]

Well I tried some sweeps on just the transformer and I'm still getting the hi pass effect, which surely means that it's broken somehow? I can't think what else it would be.

Can you recommend a transformer that I could try in place of it? The Sowter one is expensive and there's a 12 week lead time at the moment and I'd rather sort this out a bit sooner.