This is my first post here and I’m looking for some help with two problems with a valve pre-amplifier I recently put to-gether. The two problems are hum and too much gain…
Firstly, a bit of background – I’m a complete newbie to valve amplifiers and I’m doing my best to increase my knowledge by reading Morgan Jones book (Valve amplifiers) and looking at various websites. Over the last couple of months I decided to go ahead and build a pre-amplifier which was fairly quickly followed by a power amplifier.
So, the pre-amp was based on a 5687 valve as per the schematic on www.diy paradise.com which appeared to be relatively ‘simple’ to build for a newbie. It took me some time to complete but I’m new to this so no big deal! Well after testing to find all voltages were approx. correct everything looked great. However, at this stage I had no power amp so I couldn’t really test the preamp playing music.
So I decided to build a power amp. This time with some help with a schematic I built a EL34 SE power amp (using ECC83 drivers). Again, everything went ok – no problems with voltages on power amp, etc. Excitedly I coupled the pre-amp and power amp together but it was immediately obvious I had problems – loud hum.
Played some music anyway but then second problem surfaced – too much gain (volume was too loud before 9’ o clock on dial.)
Some investigation was carried out: Could play music through power amp (with Walkman CD + volume control) and there was no hum so I figured that this meant the pre-amp was at fault.
So looking at pre-amp
Using Hammond power TX (373X) and 5Y3 rectifier then 30H choke/50uf cap/30H choke/50uf Cap to give B+ of ~250V. Heaters on 5687 valve use AC so cables are twisted together.
Tried to keep signal cable away from AC and B+ as much as possible and used screened cable where possible on signal side.
Phono connectors are isolated from chassis so earthing (which proved problematic for me) is connected to chassis at one of bolts under mains transformer. This could well be wrong as I am not too sure what is best.
I only have a DMM so can’t carry out much testing to explain type of hum however it is there when volume is at minimum and is largely volume independent (although once volume pot is at ~1-2o’clock without music there is an increasing buzzing noise on both channels as well as the hum).
I’ve tried re-routing earth connections and have tried different combinations of connecting screen of signal cable to earth (e.g. at one end only, at both ends etc) but hum still persists.
Only way forward maybe to go DC on heaters (I think Bibster did this when he had similar problems with hum on a 5687 amp) but I’m not sure if my transformer (6.3VAC @ 4A) is up to spec for this change and I’m pretty sure I’d need a schematic to show me how this would be best achieved. Hopefully the hum is coming from the layout rather than heaters as the solution should be easier.
I’ve attached a photo of schematic from which it will be hopefully it will be obvious where the faults lie (to an expert eye – unlike mine).
For the gain problem I tried to remove the cathode cap but this really didn’t have any noticeable effect. Also put 150k resistor on each channel of input signal prior to Alpha (100k) volume pot which seems to have helped marginally – can now run at 9’o clock on pot without fear of damaging ‘speakers.)
Maybe I need to change the pot? Does anyone know if this is normally an issue with this 5697 pre-amp?
Any help would be gratefully appreciated - this hum problem is really driving me mad?
Firstly, a bit of background – I’m a complete newbie to valve amplifiers and I’m doing my best to increase my knowledge by reading Morgan Jones book (Valve amplifiers) and looking at various websites. Over the last couple of months I decided to go ahead and build a pre-amplifier which was fairly quickly followed by a power amplifier.
So, the pre-amp was based on a 5687 valve as per the schematic on www.diy paradise.com which appeared to be relatively ‘simple’ to build for a newbie. It took me some time to complete but I’m new to this so no big deal! Well after testing to find all voltages were approx. correct everything looked great. However, at this stage I had no power amp so I couldn’t really test the preamp playing music.
So I decided to build a power amp. This time with some help with a schematic I built a EL34 SE power amp (using ECC83 drivers). Again, everything went ok – no problems with voltages on power amp, etc. Excitedly I coupled the pre-amp and power amp together but it was immediately obvious I had problems – loud hum.
Played some music anyway but then second problem surfaced – too much gain (volume was too loud before 9’ o clock on dial.)
Some investigation was carried out: Could play music through power amp (with Walkman CD + volume control) and there was no hum so I figured that this meant the pre-amp was at fault.
So looking at pre-amp
Using Hammond power TX (373X) and 5Y3 rectifier then 30H choke/50uf cap/30H choke/50uf Cap to give B+ of ~250V. Heaters on 5687 valve use AC so cables are twisted together.
Tried to keep signal cable away from AC and B+ as much as possible and used screened cable where possible on signal side.
Phono connectors are isolated from chassis so earthing (which proved problematic for me) is connected to chassis at one of bolts under mains transformer. This could well be wrong as I am not too sure what is best.
I only have a DMM so can’t carry out much testing to explain type of hum however it is there when volume is at minimum and is largely volume independent (although once volume pot is at ~1-2o’clock without music there is an increasing buzzing noise on both channels as well as the hum).
I’ve tried re-routing earth connections and have tried different combinations of connecting screen of signal cable to earth (e.g. at one end only, at both ends etc) but hum still persists.
Only way forward maybe to go DC on heaters (I think Bibster did this when he had similar problems with hum on a 5687 amp) but I’m not sure if my transformer (6.3VAC @ 4A) is up to spec for this change and I’m pretty sure I’d need a schematic to show me how this would be best achieved. Hopefully the hum is coming from the layout rather than heaters as the solution should be easier.
I’ve attached a photo of schematic from which it will be hopefully it will be obvious where the faults lie (to an expert eye – unlike mine).
For the gain problem I tried to remove the cathode cap but this really didn’t have any noticeable effect. Also put 150k resistor on each channel of input signal prior to Alpha (100k) volume pot which seems to have helped marginally – can now run at 9’o clock on pot without fear of damaging ‘speakers.)
Maybe I need to change the pot? Does anyone know if this is normally an issue with this 5697 pre-amp?
Any help would be gratefully appreciated - this hum problem is really driving me mad?
Attachments
Regarding the hum all the information you will need is in the Bibster thread.
You can just about get enough DC out of a 6.3V winding if you use a big initial filter cap and basically just rectify it and wack the big cap on the supply. It will run the filament a little starved, but this is a good thing in my book. In my opinion a quick and dirt lash up of this might save a hell of a lot of waisted time.
Have you earthed one side of your filament winding ? Did you lift the filament to +20V as per the schematic ?
Shoog
You can just about get enough DC out of a 6.3V winding if you use a big initial filter cap and basically just rectify it and wack the big cap on the supply. It will run the filament a little starved, but this is a good thing in my book. In my opinion a quick and dirt lash up of this might save a hell of a lot of waisted time.
Have you earthed one side of your filament winding ? Did you lift the filament to +20V as per the schematic ?
Shoog
Regarding your gain problem. The diyparadise linestage doesn't have excessive gain. Therefore your gain problem lies firmly with your choice of driver in your power amp.
You have two solutions;
1-change the driver stage (the best solution)
2-put a voltage divider on the input of the amp. Use Metal Film to minimise the the sonic impact. 100K from the phono plug followed by another 100K to ground should reduce your gain.
The excessive gain of your amplifier is amplifying the hum of your line stage.
Shoog
You have two solutions;
1-change the driver stage (the best solution)
2-put a voltage divider on the input of the amp. Use Metal Film to minimise the the sonic impact. 100K from the phono plug followed by another 100K to ground should reduce your gain.
The excessive gain of your amplifier is amplifying the hum of your line stage.
Shoog
Hi Shoog,
Heater supply is earthed via CT from Transformer - so that is ok. Filament isn't lifted +20V though at the moment. Must re-read and make notes from Bibsters thread to find out how to do this.
Regarding going to DC should be able to put something together later (assuming parts are available etc).
Thanks
Colin
Heater supply is earthed via CT from Transformer - so that is ok. Filament isn't lifted +20V though at the moment. Must re-read and make notes from Bibsters thread to find out how to do this.
Regarding going to DC should be able to put something together later (assuming parts are available etc).
Thanks
Colin
Well, I'm not going to say that I buy this statement without further examination. Do you have another power amplifier in which to try your line stage with? If you don't have one can you borrow one from a neighbor or friend?
You mentioned that you have a multimeter is this correct? If so measure the output of your linestage on both the DC and AC millivolt scale and post the results.
I am suggesting an examination of both parties (linestage and SE amplifier)before you dive in and start tearing the hell out of something that may or may not be the problem. Make a few tests and come back and post your results. If you have a scope look at the output of the line stage.
It maybe that the power amp is designed for a CD level input and as such any gain will be excessive. If thats the case then the linestage could be considered wrong for the application. The diyparadise preamp is a moderate gain preamp.
Whatever the case the marriage needs some slight modification at the least, padding down the input is the most straightforward approach.
Sort out the hum first.
Shoog
Whatever the case the marriage needs some slight modification at the least, padding down the input is the most straightforward approach.
Sort out the hum first.
Shoog
Finally back after fiddling with bridge and caps all afternoon in an attempt to overcome the hum problems.
Thanks for suggestions from Shoog & burned fingers - maybe I can answer some of your points.
Regarding the hum eventually tried following combination
6.3VAC/Bridge/4700uF/0.56R(10W)/4700uF
this gave DC of ~6.15V measured at valve socket - ok!
Having only DMM available (no 'scope) unable to measure any changes in the hum with AC v's DC filament heating but my ears tell me the hum is greatly reduced.
The only problem now might be fitting the extra components into chassis.
Regarding the high gain I haven't made any changes to the power amp as suggested yet (by Shoog i.e voltage divider on input but will try this later/to-morrow). This seems better than changing the driver in the power amp at this point.
I'm a little bit worried by the marriage of the line stage amp and power amp. Whilst testing I use a Walkman CD which has a line out which I connect to line stage or I can use headphone output/vol control connect directly to the power amp. However, ultimately I want to use my main CD player which has no volume control hence the need for a line-stage. (I guess I could put a volume pot on the power amp as an alternative for CD playback making the linestage redundant but I certainly did have that in mind when I started out.)
Hopefully using voltage divider on the input of the power amp will solve the gain problems.
BTW I'm using 3 wire grounded power cord on both amps.
Colin.
Thanks for suggestions from Shoog & burned fingers - maybe I can answer some of your points.
Regarding the hum eventually tried following combination
6.3VAC/Bridge/4700uF/0.56R(10W)/4700uF
this gave DC of ~6.15V measured at valve socket - ok!
Having only DMM available (no 'scope) unable to measure any changes in the hum with AC v's DC filament heating but my ears tell me the hum is greatly reduced.
The only problem now might be fitting the extra components into chassis.
Regarding the high gain I haven't made any changes to the power amp as suggested yet (by Shoog i.e voltage divider on input but will try this later/to-morrow). This seems better than changing the driver in the power amp at this point.
I'm a little bit worried by the marriage of the line stage amp and power amp. Whilst testing I use a Walkman CD which has a line out which I connect to line stage or I can use headphone output/vol control connect directly to the power amp. However, ultimately I want to use my main CD player which has no volume control hence the need for a line-stage. (I guess I could put a volume pot on the power amp as an alternative for CD playback making the linestage redundant but I certainly did have that in mind when I started out.)
Hopefully using voltage divider on the input of the power amp will solve the gain problems.
BTW I'm using 3 wire grounded power cord on both amps.
Colin.
Glad to hear your've made real progress with the hum. I would be inclined to make your second capacitor at least 20,000uf to get the best DC. Because you are restricted to purely capacitive filtering, and the current draw is relatively high - you need as much as possible. You only need 10V ratings so that should still be relatively small.
A volume pot is basically a variable voltage divider but with more flexability. Still I would say that the voltage divider with MF should sound better so if you can get it working to your satisfaction thats the way to go. Inevitably you will experience a slight loss of detail. The secondary advantage will be that because the hum is preamp induced, you should kill most of the remaining hum.
Keep us posted.
Shoog
A volume pot is basically a variable voltage divider but with more flexability. Still I would say that the voltage divider with MF should sound better so if you can get it working to your satisfaction thats the way to go. Inevitably you will experience a slight loss of detail. The secondary advantage will be that because the hum is preamp induced, you should kill most of the remaining hum.
Keep us posted.
Shoog
I'll try that as well, once I'll have some time to fiddle and re-case my linestage. Hope it'll remove the last bit of hum.Shoog said:
I can't see why a volume pot is more flexible than a variable voltage divider...Shoog said:
You could try quick'n'dirty el-cheapo pot, find the right 'position', and then measure, round it up to 'normal' (E12, E96) resistor values, set it that way, listen, approve, and replace by MF resistors....
Good luck, Bibster
Oh, and by the way, I've lifted my heaters by means of a 10:1 voltage divider (220K:22K) off the B+
Shoog,
I managed to get them working, they're running quite hot though.
But, as they keep falling out of the cabinet.... (Yes that's right, physically falling out)
Sounds great, I'd say: Way more quiet, straighter sound, and way better channel sep.
But, as I said: they keep falling....
Need to do a new case, with a more logical lay-out: There just wires crossing everywhere, low level signal stuff all over the place, stuck to the recifier, heaterwires making love to PS chokes, output wires // to AC lines for miles etc. etc. etc.
But, my trouble and strife forbits me to make a mess in the garage for now (She just cleaned it, you see...) so I'll start thinking about the new case once I'm back from my (well deserved) vacation in Sneek!
Paul
I managed to get them working, they're running quite hot though.
But, as they keep falling out of the cabinet.... (Yes that's right, physically falling out)
Sounds great, I'd say: Way more quiet, straighter sound, and way better channel sep.
But, as I said: they keep falling....
Need to do a new case, with a more logical lay-out: There just wires crossing everywhere, low level signal stuff all over the place, stuck to the recifier, heaterwires making love to PS chokes, output wires // to AC lines for miles etc. etc. etc.
But, my trouble and strife forbits me to make a mess in the garage for now (She just cleaned it, you see...) so I'll start thinking about the new case once I'm back from my (well deserved) vacation in Sneek!
Paul
Hi Shoog,
Thanks for your recommendation of second capacitor of at least 20,000uF for the DC filament supply. Unfortunately, I don't have one to hand so I can't experiment further with this. I do have 2x 10000uF (63V) which could be paralleled but there a bit large really. My local electronics supplier doesn't stock 20,000uF (10V) so I need to do some searching on the Web to locate some!
I have a L317 voltage regulator that at the bottom of my box of spares at the moment and I've been contemplating the possibility of incorporating this in the DC supply. But correct me if I'm wrong but i think the 6.3VAC transformer even after rectifing using a SS bridge will still be at too low a voltage for the L317 to ultimately produce a regulated ~6.3VDC for the 5687 tube?
I've also tried your suggestion of a voltage divider on the signal input to the power amp using two 100k resistors. The gain has definitely dropped (as expected) but of course without an oscilloscope I can't really quantify this. However, this looks really promising now - many thanks.
So, just need to finalise components for the DC supply to the heaters then everything should be ok.
BTW, Bibster (Paul) thanks for your support. Noted your comment regarding 'I've lifted my heaters by means of a 10:1 voltage divider (220K:22K) off the B+'. Can you advise whether I need to do this as well? At the moment my 6.3VAC supply for the 5687 filaments from the TX is CT and this is already grounded so do I stick with this or change to your set-up. I'm pretty clueless about this!
Colin.
Thanks for your recommendation of second capacitor of at least 20,000uF for the DC filament supply. Unfortunately, I don't have one to hand so I can't experiment further with this. I do have 2x 10000uF (63V) which could be paralleled but there a bit large really. My local electronics supplier doesn't stock 20,000uF (10V) so I need to do some searching on the Web to locate some!
I have a L317 voltage regulator that at the bottom of my box of spares at the moment and I've been contemplating the possibility of incorporating this in the DC supply. But correct me if I'm wrong but i think the 6.3VAC transformer even after rectifing using a SS bridge will still be at too low a voltage for the L317 to ultimately produce a regulated ~6.3VDC for the 5687 tube?
I've also tried your suggestion of a voltage divider on the signal input to the power amp using two 100k resistors. The gain has definitely dropped (as expected) but of course without an oscilloscope I can't really quantify this. However, this looks really promising now - many thanks.
So, just need to finalise components for the DC supply to the heaters then everything should be ok.
BTW, Bibster (Paul) thanks for your support. Noted your comment regarding 'I've lifted my heaters by means of a 10:1 voltage divider (220K:22K) off the B+'. Can you advise whether I need to do this as well? At the moment my 6.3VAC supply for the 5687 filaments from the TX is CT and this is already grounded so do I stick with this or change to your set-up. I'm pretty clueless about this!
Colin.
You won't have enough for a reg, but you don't need it. If you use Shottky's (1N 5822 or equivilant), you'll have more voltage and less noise. With a filament winding I did that and 10mf/1r/10mf/1r/10mf. It models to about 3mv ripple and give about 6.3 volts (assuming 7volts unloaded). No need to lift the DC supply.
Sheldon
Sheldon
Sheldons looks like an excellent plan.
I would be tempted to lash up the 10,000uf 63V just to see if you get any improvement. If not then don't bother using any more capacitance than you already have. If you have any dead PC power supplies there is usually a good range of quality 10V caps in it which can be salvaged.
Shoog
I would be tempted to lash up the 10,000uf 63V just to see if you get any improvement. If not then don't bother using any more capacitance than you already have. If you have any dead PC power supplies there is usually a good range of quality 10V caps in it which can be salvaged.
Shoog
Hi there,
I just thought I would tell you my little experience with using Metal Film resistors on a voltage divider of my own amp. I tried as I suggested and found the sound turned glassy and bright. I decided to replace the series resistor with a carbon film whilst leaving the Metal Film as the shunt. This has taken the edge off the overall presentation.
Another example of suck it and see !
Shoog
I just thought I would tell you my little experience with using Metal Film resistors on a voltage divider of my own amp. I tried as I suggested and found the sound turned glassy and bright. I decided to replace the series resistor with a carbon film whilst leaving the Metal Film as the shunt. This has taken the edge off the overall presentation.
Another example of suck it and see !
Shoog
Hi Shoog,
Sorry for the lack of updates regarding the hum on the 5687 pre-amp but I've been waiting for some 10,000uF caps to arrive by post. I finally got them on Friday but I have been away all weekend in the Lake District with the weather being so good for once so I've had no chance to make any progress.
I'm planning on implementing Sheldon's suggestion in the next few days to finally cure the hum. Once that's done I'll investigate your recommendation regarding metal film and carbon film resistors. I've already tried your suggestion of voltage divider on the power amp input but I used metal film resistors only and I was solely concentrating on reducing the gain (which happened)and not so much the overall sound.
Anyway thanks for the suggestion. Will keep you updated.
Colin.
Sorry for the lack of updates regarding the hum on the 5687 pre-amp but I've been waiting for some 10,000uF caps to arrive by post. I finally got them on Friday but I have been away all weekend in the Lake District with the weather being so good for once so I've had no chance to make any progress.
I'm planning on implementing Sheldon's suggestion in the next few days to finally cure the hum. Once that's done I'll investigate your recommendation regarding metal film and carbon film resistors. I've already tried your suggestion of voltage divider on the power amp input but I used metal film resistors only and I was solely concentrating on reducing the gain (which happened)and not so much the overall sound.
Anyway thanks for the suggestion. Will keep you updated.
Colin.
colinB said:
That will make a quiet filament supply, but I can see maybe one other issue. From your picture, the chassis looks to be steel. You could be getting some inductive hum from your main transformer. If so, it would be line frequency hum. Do a simple test. Unbolt the transformer from the chassis and place a piece of insulating material (about 2mm thick) between transformer and chassis. Try perf board if you have some laying around. If that helps, you can find something pretty later. Wouldn't hurt on the chokes either.
Sheldon
bibster,
I built a very similar preamp and experienced a hum issue. I built the "Simple 5687 linestage". I have (2) sugestions:
1. If you have a volume pot, ensure that it is isolated from the chassis.
2. Consider micrphonics. I noticed if I used a military grade tube, the hum did decrease.
Cheers
Stuben
I built a very similar preamp and experienced a hum issue. I built the "Simple 5687 linestage". I have (2) sugestions:
1. If you have a volume pot, ensure that it is isolated from the chassis.
2. Consider micrphonics. I noticed if I used a military grade tube, the hum did decrease.
Cheers
Stuben
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.