Geloso G1/1040 Amp repair

Burns, Selmer, WEM, and more

Geloso G1/1040 Amp repair

Postby zjokka on Sun Sep 11, 2016 9:15 pm

Hi all,

Kind of new here. I saw this video and then forgot all about buying a amp kit to build...


So I went out and found a Geloso 3215 with as only sound the crackling volume pot... until I changed the mains switch from 240V to 220V. Now it's growling like a Testa Rossa when played, but else dead silent. To add to all this happiness it had all the original factory tubes and they are fine. It seems like I'll never sell this amp, so I might just sell the original tubes to a collector and buy two more Geloso's....

Image

The G1040 is not giving out any sound at all... but at least the reason why is clear. Filaments are not burning on the 2 12AX7 tubes, they are not getting any warmer. I am not an electronics expert, but have been documenting myself like hell, but most documentation about guitar amps are based off Fender designed circuits. Very strange to find the two first tubes' filaments not only powered by DC, but negative DC.

Voltage measured on the non working heaters:
12AX7(1) 4 = 0V; 5 = -5.85V
12AX7(2) 4 = -5.4V; 5 = -12.5V

Is this the problem? Should I be seeing -6V here to turn on the light?

I have -26V coming off the rectifier and have already replaced the 800µF/50V capacitor with 2 parallel 470µF/30V's. Is this enough?
Before the high voltage rectifier is only see 220VAC instead of the 330V indicated on the transformer.... after the BY126 rectifier I see 245V only, which seems low. One more symptom: the grids on the second tube are slightly positive +0.03 and +0.04 as well as the cathodes at 0.5V and 0.77V.

The 100R 1W resistor has seens some action, the legs have corroded green coloring, but it measures out fine. Change caps? Replace rectifier? Give up on the PT?

It is clear that the second 12AX7 is where the problem resides. The current limiter is very slighly lit when I turn it on (after the caps are charged), but lighting up a whole lot more when I turn on B+. When I then remove the second 12ax7, the light grows considerably dimmer again.

Thanks in advance for any pointers.
zjokka
Peasant
 
Posts: 13
Joined: Sun Sep 11, 2016 3:37 pm

Re: Geloso G1/1040 Amp repair

Postby trobbins on Thu Sep 15, 2016 7:39 am

Using the bias for heater DC supply was a convenience of design. The heaters act as a CRCR filter load, so the heater connecting to 0V is better filtered and would be the input stages.

As your voltages are low on the heaters, and you've replaced the filter caps (hopefully with correct polarity), then check the bridge rectifier diodes (swap out), - it looks like just a bad diode causing half-wave rectification, and hence low output bias voltage under load. The low bias would cause stress on the output stage valves, and load down the B+ severely. The heaters are unlikely to have the same DCV across them at very low power.

Check for correct bias and 12AX7 heater operation with the standby off. Check heater voltages on all valves - set the AC mains level tap to give nominal heater voltage.

The rest of the amp probably needs a good check for output valve bias currents (insert a cathode sense resistor is prudent), and that all the other valves have acceptable plate voltages (the input stages are grid leak biased, so plate voltage is your only guide).
trobbins
Court Jester
 
Posts: 74
Joined: Wed Oct 14, 2009 7:30 am
Location: Melbourne, OZ

Re: Geloso G1/1040 Amp repair

Postby zjokka on Tue Sep 27, 2016 9:23 pm

Thanks T,

Sorry about the late reply, have been busy at work, in the meantime also brought another Geloso (3215) to live and just loving it. Also doing final debugging on a 5E3 that I built over the summer.

Ok, we are talking about replacing the packaged bridge rectifier, not the one with the diodes visible. This is it:
Image

I only replaced one of the two capactitors in that line, maybe I should do that first.
But my main question is now: can I use like 1N400x diodes? Have a bunch of different types lying around...
zjokka
Peasant
 
Posts: 13
Joined: Sun Sep 11, 2016 3:37 pm

Re: Geloso G1/1040 Amp repair

Postby stratele52 on Tue Sep 27, 2016 10:41 pm

1N400X diode are only 1 amp diode. I'll use higher amps diode at 600 to 1000 volts PIV

Why not a one piece bridge rectifier ?
stratele52
Knight
 
Posts: 1904
Joined: Sat Aug 21, 2010 9:35 am
Location: Québec province, Canada

Re: Geloso G1/1040 Amp repair

Postby trobbins on Wed Sep 28, 2016 1:23 am

The selenium bridge being replaced is for the bias and input stage heater supply - it will only have continuous current requirement of about 150mA in that circuit as the 12AX7 heaters are in series, and using 1A nominal diodes such as 1N4002 or upwards in voltage rating would be fine, even for the heater in-rush.

The bias and heater operating levels will likely need some adjustment (higher series resistances) as the the silicon diodes will have lower voltage drop, so the raw DC voltage will be somewhat higher.
trobbins
Court Jester
 
Posts: 74
Joined: Wed Oct 14, 2009 7:30 am
Location: Melbourne, OZ

Re: Geloso G1/1040 Amp repair

Postby stratele52 on Wed Sep 28, 2016 7:55 am

trobbins wrote:The selenium bridge being replaced is for the bias and input stage heater supply - it will only have continuous current requirement of about 150mA in that circuit as the 12AX7 heaters are in series, and using 1A nominal diodes such as 1N4002 or upwards in voltage rating would be fine, even for the heater in-rush.

.


I did not read it was for bias,
stratele52
Knight
 
Posts: 1904
Joined: Sat Aug 21, 2010 9:35 am
Location: Québec province, Canada

Re: Geloso G1/1040 Amp repair

Postby Racing on Tue Oct 18, 2016 7:38 am

When we take your measured voltage specs into account it is reasonable that something is loading that PT in a rather massive manner.
Check to see which part of it.
The mere fact that you´re low on B+ is telling something. Start by checking the fuse rating of that circuit for starters. Then yank the powertubes,the EL-34´s,and try again. The 1040 is what we´d call a low voltage amp seing it´s EL-34´s but..not THAT low.
If memory serves me you should register around the just shy of 400V mark or so under load. If B+ is still down with powertubes out of the equation my next bet would be the onboard smoothing capacitors. Should really be checked on an LCR meter for ESR and capacitance.

The old BY series diodes in there should suffice and then some but if the issue persists after those two above has been cleared out that´d be next thing in order to check.
Also be aware,shadetree solution,that a simple DC resistence check from various points in the DC rail vs ground can tell a lot too.

The entire issue of lack of heat for the two upmost preamp tubes is a rather minor matter IMO.

First up find out what is loading that PT down.
Racing
Squire
 
Posts: 749
Joined: Wed Dec 21, 2011 11:07 am

Re: Geloso G1/1040 Amp repair

Postby zjokka on Fri Oct 21, 2016 11:48 am

Racing wrote:First up find out what is loading that PT down.

Ok, I finally got around to getting some hard figures.

These are the voltages without the tubes in place: I see 360V after the rectifier, so this is much better...
Image

Here are the voltages with the tubes in place:
Image

So next up: checking all the smoothing caps.... don't have 16uFs, will have to put in 22uF's
Image
zjokka
Peasant
 
Posts: 13
Joined: Sun Sep 11, 2016 3:37 pm

Re: Geloso G1/1040 Amp repair

Postby Racing on Sun Oct 23, 2016 11:28 am

Something is amiss,and you need to figure out why.

First up the quoted 220VAC out of the PT is to small a number. A 1040 should see an easy 350VDC under load meaning that your unloaded numbers AC are wrong.
Start by checking DC resistence of secondary PT winding. This should read a relatively low number in Ohms. Ditto for primary. Should read along the lines of a few Ohms.

If that checks out....
My bet is that something is pulling LOTS of current here. Check DC resistence between voltage rail and chassis ground with amp off. That´s a good starting point at least.
If tubes are out there should be no appreciable drop in voltage,dropresistors or not,for the entire amp. Hence...

In turn 220VAC and 365VDC don´t add up either. As you rectify AC voltage you should see a DC voltage around the 1,3 mark or so-under load. I appreciate that those are unloaded numbers,they´re still not up to snuff.

I´d say disconnect all of it from the rail,down to the OT. Install fresh diodes and smoothing cap for first hit. With ONLY that in the equation check both AC and DC voltage. Make sure the e-cap has ample voltage capacity.
Then add the OT,only. Voltage should remain the same. Check DC resistence of OT from centertap to each side and write that number down (of course with amp turned off).
Check that your bias voltage for g1 is about on par,meaning +30VDC negative and then some.
Then add voltage to g2. In this manner the powertubes ONLY should be up and running. Voltage should decrease,WITHIN REASON. Again check voltage vs g1,and do so under load. It should remain around them +30VDC negative. If not,lift one end of offending coupling cap vs the phaseinverter and try again.
Add the smoothing capacitor,fresh such,as you go.
Check bias by measuring voltage drop from OT centertap to respective powertubes pin 3. That number divided by the X number of Ohms you measured from CT to each side will give you the load in mA. Now multiply that by B+ voltage measured at output transformers centertap.
U*I=P
That way you get anode loss in idle. Aka bias.

Also make sure that the grounding of the first hit electrolyte meets up with bridge negative first thing. NOTHING else is to be let to that point. This is what´s known as "the golden rule",and is a function of one of Kirchoffs laws. Google need be.
Racing
Squire
 
Posts: 749
Joined: Wed Dec 21, 2011 11:07 am

Re: Geloso G1/1040 Amp repair

Postby trobbins on Mon Oct 24, 2016 6:51 am

Z, your measurements on the first schematic are difficult to read, as there appears to be some readings with a "double underline" ???

Can your meter read ACV, especially the transformer secondary ACV values when no valves are inserted. You need to confirm that your heater ACV is acceptable (ie. 6.3V +/-0.3V when the primary tapping is set to appropriate value). At the moment your -22V bias voltage seems low for no loading given that the bias needs 25V nominal when loaded, and then you show a -31.7V reading after the extra filter (and -26V for your latter measurements with heaters) ??? You may need to recheck those readings - do you have another meter to cross-check with ?

Your next test should have been without output valves, but with input stage valves, and with standby switch off. That would confirm that the bias voltage is acceptable when it is loaded with input stage heaters.

The your next test should be with the standby on, but again no output valves, as that would confirm that all the stages up to the splitter are nominal.

It's hard to tell if your 2nd set of measurements have the splitter and output stage valves in circuit??
trobbins
Court Jester
 
Posts: 74
Joined: Wed Oct 14, 2009 7:30 am
Location: Melbourne, OZ

Re: Geloso G1/1040 Amp repair

Postby zjokka on Sat Oct 29, 2016 8:51 pm

Thanks for the replies, really had little time to work on this the last week(s), work is a pain. Also some other shit not worth mentioning.
Racing wrote:First up the quoted 220VAC out of the PT is to small a number. A 1040 should see an easy 350VDC under load meaning that your unloaded numbers AC are wrong.

Today I tried to fire this beast up. First off, I had the speaker wiring completely wrong for 4 ohms. Had one speaker line attached to the chassis :Doh:
My only working Geloso I resuscitated by playing with the input voltage selector, which was set wrong.
Also with the 1040, there is 220 or 240 -- while we have 230V here in Belgium. Anyway, I tried 240V, fired it up ...nothing happened, turned out the fuse on the main selector blew. So no voltage testing today until I get a new fuse.
Start by checking DC resistence of secondary PT winding. This should read a relatively low number in Ohms. Ditto for primary. Should read along the lines of a few Ohms.

Secondary winding resistances:
HV: 17 ohm
6.3V: 1.9 ohm
40V: 1.7 ohm
If that checks out....
My bet is that something is pulling LOTS of current here. Check DC resistence between voltage rail and chassis ground with amp off. That´s a good starting point at least.If tubes are out there should be no appreciable drop in voltage,dropresistors or not,for the entire amp. Hence...

Not exactly sure what you mean by voltage rail. At the 16uF filter caps there is infinite resistance to ground.
I did desolder the 16uF and the 40uF, they were good, so reinstalled them.
Will report back after getting new fuse and measurements.
zjokka
Peasant
 
Posts: 13
Joined: Sun Sep 11, 2016 3:37 pm

Re: Geloso G1/1040 Amp repair

Postby zjokka on Mon Oct 31, 2016 1:16 am

Ok, I found a 2.5A fuse which fitted
Image

Replaced 500uF/50V capacitor connected to the 100Ω/4W:
Image

I did not have proper replacements for the 40uF filter caps -- checked them, they should be ok.
Image

Connected a 4Ω speaker to the correct output. And this beast started talking.
I have current limited light bulb connected: when I turn the amp in standby the bulb lights up and dies to almost nothing while the caps are charging. Filament of the lamp is lit slightly, but when I turn on the HV, it lights up again. I have 340V DC hitting the first filter cap. I will make further measurements tomorrow.

So I am losing AC somewhere? or couuld it be DC? Will put brand new tubes in the amp tomorrow.

Still, the amp is working near to flawlessly... I have no means to test it properly, it's too late tonight, will get back to that tomorrow, but it is dead silent. Dead silent with the emphasis on dead... if this was a Fender amp you would swear it was dead. Only tested it with an acoustic with piezo I have lying around in the veranda here. Tone controls are really hifi like, no surprise, but for an acoustic fitted with a K&K Pure Mini, this would be almost giggable.
zjokka
Peasant
 
Posts: 13
Joined: Sun Sep 11, 2016 3:37 pm

Re: Geloso G1/1040 Amp repair

Postby trobbins on Mon Oct 31, 2016 11:31 am

Imho, you shouldn't turn on the HV if the bias voltage isn't correct - that would just stress your output valves for no good reason. Was the bias ok, and the input stage heater voltage ok with standby off? If not, then you need to fix that problem first.
trobbins
Court Jester
 
Posts: 74
Joined: Wed Oct 14, 2009 7:30 am
Location: Melbourne, OZ

Re: Geloso G1/1040 Amp repair

Postby zjokka on Mon Oct 31, 2016 6:03 pm

trobbins wrote:Imho, you shouldn't turn on the HV if the bias voltage isn't correct - that would just stress your output valves for no good reason. Was the bias ok, and the input stage heater voltage ok with standby off? If not, then you need to fix that problem first.

Racing wrote:Check that your bias voltage for g1 is about on par,meaning +30VDC negative and then some.
Then add voltage to g2. In this manner the powertubes ONLY should be up and running. Voltage should decrease,WITHIN REASON. Again check voltage vs g1,and do so under load. It should remain around them +30VDC negative. If not,lift one end of offending coupling cap vs the phaseinverter and try again

-31VDC! :jam:
zjokka
Peasant
 
Posts: 13
Joined: Sun Sep 11, 2016 3:37 pm

Re: Geloso G1/1040 Amp repair

Postby trobbins on Tue Nov 01, 2016 12:52 am

Are the heater voltages ok that are powered from the bias?

PS. 31V bias reading indicates 39V from bias supply. Nominal 150mA through heaters would require 12.6+12.6+100x0.15 = 40V.

Best to insert cathode current sense resistor for each EL34, then you can easily assess if each valve is ok, and whether you have nominal balanced valves, and can calculate valve dissipation to see whether the bias voltage needs to be tweaked up/down a bit by changing the 8K2.

You may want to disconnect the feedback whilst setting up the bias - just in case there is some oscillation happening in the background that you aren't aware of.
trobbins
Court Jester
 
Posts: 74
Joined: Wed Oct 14, 2009 7:30 am
Location: Melbourne, OZ

Next

Return to Miscellaneous British Amplifiers

Who is online

Users browsing this forum: No registered users and 1 guest